Commit
·
7b693ed
1
Parent(s):
65797a2
init
Browse files- README.md +15 -0
- added_tokens.json +5 -0
- config.json +861 -0
- generation_config.json +16 -0
- merges.txt +0 -0
- model-00001-of-00004.safetensors +3 -0
- model-00002-of-00004.safetensors +3 -0
- model-00003-of-00004.safetensors +3 -0
- model-00004-of-00004.safetensors +3 -0
- model.safetensors.index.json +0 -0
- special_tokens_map.json +20 -0
- tokenizer.json +0 -0
- tokenizer_config.json +44 -0
- trainer_state.json +2219 -0
- training_args.bin +3 -0
- vocab.json +0 -0
README.md
CHANGED
|
@@ -1,3 +1,18 @@
|
|
| 1 |
---
|
| 2 |
license: mit
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: mit
|
| 3 |
---
|
| 4 |
+
|
| 5 |
+
# M4-Auio-7B
|
| 6 |
+
|
| 7 |
+
Enhancing Interactive Capabilities in VideoLLM
|
| 8 |
+
|
| 9 |
+
M4-Audio-7B is an extension of [LongVA-7B](https://github.com/EvolvingLMMs-Lab/LongVA), further trained using the [M4-IT](https://huggingface.co/datasets/ColorfulAI/M4-IT) dataset, which comprises 9,963 visual-audio instruction tuning instances. This training was conducted without any special modifications to the existing training pipeline.
|
| 10 |
+
|
| 11 |
+
|
| 12 |
+
## Usage
|
| 13 |
+
|
| 14 |
+
|
| 15 |
+
|
| 16 |
+
|
| 17 |
+
For more information about the interaction inference pipeline, please visit the [M4 GitHub repository](https://github.com/patrick-tssn/M4).
|
| 18 |
+
|
added_tokens.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<|endoftext|>": 151643,
|
| 3 |
+
"<|im_end|>": 151645,
|
| 4 |
+
"<|im_start|>": 151644
|
| 5 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,861 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "checkpoints/longva7b-qwen2-voiceassistant",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LlavaQwenForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 151643,
|
| 8 |
+
"eos_token_id": 151645,
|
| 9 |
+
"hidden_act": "silu",
|
| 10 |
+
"hidden_size": 3584,
|
| 11 |
+
"image_aspect_ratio": "anyres",
|
| 12 |
+
"image_crop_resolution": null,
|
| 13 |
+
"image_grid_pinpoints": [
|
| 14 |
+
[
|
| 15 |
+
336,
|
| 16 |
+
672
|
| 17 |
+
],
|
| 18 |
+
[
|
| 19 |
+
336,
|
| 20 |
+
1008
|
| 21 |
+
],
|
| 22 |
+
[
|
| 23 |
+
336,
|
| 24 |
+
1344
|
| 25 |
+
],
|
| 26 |
+
[
|
| 27 |
+
336,
|
| 28 |
+
1680
|
| 29 |
+
],
|
| 30 |
+
[
|
| 31 |
+
336,
|
| 32 |
+
2016
|
| 33 |
+
],
|
| 34 |
+
[
|
| 35 |
+
336,
|
| 36 |
+
2352
|
| 37 |
+
],
|
| 38 |
+
[
|
| 39 |
+
336,
|
| 40 |
+
2688
|
| 41 |
+
],
|
| 42 |
+
[
|
| 43 |
+
336,
|
| 44 |
+
3024
|
| 45 |
+
],
|
| 46 |
+
[
|
| 47 |
+
336,
|
| 48 |
+
3360
|
| 49 |
+
],
|
| 50 |
+
[
|
| 51 |
+
336,
|
| 52 |
+
3696
|
| 53 |
+
],
|
| 54 |
+
[
|
| 55 |
+
336,
|
| 56 |
+
4032
|
| 57 |
+
],
|
| 58 |
+
[
|
| 59 |
+
336,
|
| 60 |
+
4368
|
| 61 |
+
],
|
| 62 |
+
[
|
| 63 |
+
336,
|
| 64 |
+
4704
|
| 65 |
+
],
|
| 66 |
+
[
|
| 67 |
+
336,
|
| 68 |
+
5040
|
| 69 |
+
],
|
| 70 |
+
[
|
| 71 |
+
336,
|
| 72 |
+
5376
|
| 73 |
+
],
|
| 74 |
+
[
|
| 75 |
+
336,
|
| 76 |
+
5712
|
| 77 |
+
],
|
| 78 |
+
[
|
| 79 |
+
336,
|
| 80 |
+
6048
|
| 81 |
+
],
|
| 82 |
+
[
|
| 83 |
+
336,
|
| 84 |
+
6384
|
| 85 |
+
],
|
| 86 |
+
[
|
| 87 |
+
336,
|
| 88 |
+
6720
|
| 89 |
+
],
|
| 90 |
+
[
|
| 91 |
+
336,
|
| 92 |
+
7056
|
| 93 |
+
],
|
| 94 |
+
[
|
| 95 |
+
336,
|
| 96 |
+
7392
|
| 97 |
+
],
|
| 98 |
+
[
|
| 99 |
+
336,
|
| 100 |
+
7728
|
| 101 |
+
],
|
| 102 |
+
[
|
| 103 |
+
336,
|
| 104 |
+
8064
|
| 105 |
+
],
|
| 106 |
+
[
|
| 107 |
+
336,
|
| 108 |
+
8400
|
| 109 |
+
],
|
| 110 |
+
[
|
| 111 |
+
336,
|
| 112 |
+
8736
|
| 113 |
+
],
|
| 114 |
+
[
|
| 115 |
+
336,
|
| 116 |
+
9072
|
| 117 |
+
],
|
| 118 |
+
[
|
| 119 |
+
336,
|
| 120 |
+
9408
|
| 121 |
+
],
|
| 122 |
+
[
|
| 123 |
+
336,
|
| 124 |
+
9744
|
| 125 |
+
],
|
| 126 |
+
[
|
| 127 |
+
336,
|
| 128 |
+
10080
|
| 129 |
+
],
|
| 130 |
+
[
|
| 131 |
+
336,
|
| 132 |
+
10416
|
| 133 |
+
],
|
| 134 |
+
[
|
| 135 |
+
336,
|
| 136 |
+
10752
|
| 137 |
+
],
|
| 138 |
+
[
|
| 139 |
+
336,
|
| 140 |
+
11088
|
| 141 |
+
],
|
| 142 |
+
[
|
| 143 |
+
336,
|
| 144 |
+
11424
|
| 145 |
+
],
|
| 146 |
+
[
|
| 147 |
+
336,
|
| 148 |
+
11760
|
| 149 |
+
],
|
| 150 |
+
[
|
| 151 |
+
336,
|
| 152 |
+
12096
|
| 153 |
+
],
|
| 154 |
+
[
|
| 155 |
+
336,
|
| 156 |
+
12432
|
| 157 |
+
],
|
| 158 |
+
[
|
| 159 |
+
336,
|
| 160 |
+
12768
|
| 161 |
+
],
|
| 162 |
+
[
|
| 163 |
+
336,
|
| 164 |
+
13104
|
| 165 |
+
],
|
| 166 |
+
[
|
| 167 |
+
336,
|
| 168 |
+
13440
|
| 169 |
+
],
|
| 170 |
+
[
|
| 171 |
+
336,
|
| 172 |
+
13776
|
| 173 |
+
],
|
| 174 |
+
[
|
| 175 |
+
336,
|
| 176 |
+
14112
|
| 177 |
+
],
|
| 178 |
+
[
|
| 179 |
+
336,
|
| 180 |
+
14448
|
| 181 |
+
],
|
| 182 |
+
[
|
| 183 |
+
336,
|
| 184 |
+
14784
|
| 185 |
+
],
|
| 186 |
+
[
|
| 187 |
+
336,
|
| 188 |
+
15120
|
| 189 |
+
],
|
| 190 |
+
[
|
| 191 |
+
336,
|
| 192 |
+
15456
|
| 193 |
+
],
|
| 194 |
+
[
|
| 195 |
+
336,
|
| 196 |
+
15792
|
| 197 |
+
],
|
| 198 |
+
[
|
| 199 |
+
336,
|
| 200 |
+
16128
|
| 201 |
+
],
|
| 202 |
+
[
|
| 203 |
+
336,
|
| 204 |
+
16464
|
| 205 |
+
],
|
| 206 |
+
[
|
| 207 |
+
672,
|
| 208 |
+
336
|
| 209 |
+
],
|
| 210 |
+
[
|
| 211 |
+
672,
|
| 212 |
+
672
|
| 213 |
+
],
|
| 214 |
+
[
|
| 215 |
+
672,
|
| 216 |
+
1008
|
| 217 |
+
],
|
| 218 |
+
[
|
| 219 |
+
672,
|
| 220 |
+
1344
|
| 221 |
+
],
|
| 222 |
+
[
|
| 223 |
+
672,
|
| 224 |
+
1680
|
| 225 |
+
],
|
| 226 |
+
[
|
| 227 |
+
672,
|
| 228 |
+
2016
|
| 229 |
+
],
|
| 230 |
+
[
|
| 231 |
+
672,
|
| 232 |
+
2352
|
| 233 |
+
],
|
| 234 |
+
[
|
| 235 |
+
672,
|
| 236 |
+
2688
|
| 237 |
+
],
|
| 238 |
+
[
|
| 239 |
+
672,
|
| 240 |
+
3024
|
| 241 |
+
],
|
| 242 |
+
[
|
| 243 |
+
672,
|
| 244 |
+
3360
|
| 245 |
+
],
|
| 246 |
+
[
|
| 247 |
+
672,
|
| 248 |
+
3696
|
| 249 |
+
],
|
| 250 |
+
[
|
| 251 |
+
672,
|
| 252 |
+
4032
|
| 253 |
+
],
|
| 254 |
+
[
|
| 255 |
+
672,
|
| 256 |
+
4368
|
| 257 |
+
],
|
| 258 |
+
[
|
| 259 |
+
672,
|
| 260 |
+
4704
|
| 261 |
+
],
|
| 262 |
+
[
|
| 263 |
+
672,
|
| 264 |
+
5040
|
| 265 |
+
],
|
| 266 |
+
[
|
| 267 |
+
672,
|
| 268 |
+
5376
|
| 269 |
+
],
|
| 270 |
+
[
|
| 271 |
+
672,
|
| 272 |
+
5712
|
| 273 |
+
],
|
| 274 |
+
[
|
| 275 |
+
672,
|
| 276 |
+
6048
|
| 277 |
+
],
|
| 278 |
+
[
|
| 279 |
+
672,
|
| 280 |
+
6384
|
| 281 |
+
],
|
| 282 |
+
[
|
| 283 |
+
672,
|
| 284 |
+
6720
|
| 285 |
+
],
|
| 286 |
+
[
|
| 287 |
+
672,
|
| 288 |
+
7056
|
| 289 |
+
],
|
| 290 |
+
[
|
| 291 |
+
672,
|
| 292 |
+
7392
|
| 293 |
+
],
|
| 294 |
+
[
|
| 295 |
+
672,
|
| 296 |
+
7728
|
| 297 |
+
],
|
| 298 |
+
[
|
| 299 |
+
672,
|
| 300 |
+
8064
|
| 301 |
+
],
|
| 302 |
+
[
|
| 303 |
+
1008,
|
| 304 |
+
336
|
| 305 |
+
],
|
| 306 |
+
[
|
| 307 |
+
1008,
|
| 308 |
+
672
|
| 309 |
+
],
|
| 310 |
+
[
|
| 311 |
+
1008,
|
| 312 |
+
1008
|
| 313 |
+
],
|
| 314 |
+
[
|
| 315 |
+
1008,
|
| 316 |
+
1344
|
| 317 |
+
],
|
| 318 |
+
[
|
| 319 |
+
1008,
|
| 320 |
+
1680
|
| 321 |
+
],
|
| 322 |
+
[
|
| 323 |
+
1008,
|
| 324 |
+
2016
|
| 325 |
+
],
|
| 326 |
+
[
|
| 327 |
+
1008,
|
| 328 |
+
2352
|
| 329 |
+
],
|
| 330 |
+
[
|
| 331 |
+
1008,
|
| 332 |
+
2688
|
| 333 |
+
],
|
| 334 |
+
[
|
| 335 |
+
1008,
|
| 336 |
+
3024
|
| 337 |
+
],
|
| 338 |
+
[
|
| 339 |
+
1008,
|
| 340 |
+
3360
|
| 341 |
+
],
|
| 342 |
+
[
|
| 343 |
+
1008,
|
| 344 |
+
3696
|
| 345 |
+
],
|
| 346 |
+
[
|
| 347 |
+
1008,
|
| 348 |
+
4032
|
| 349 |
+
],
|
| 350 |
+
[
|
| 351 |
+
1008,
|
| 352 |
+
4368
|
| 353 |
+
],
|
| 354 |
+
[
|
| 355 |
+
1008,
|
| 356 |
+
4704
|
| 357 |
+
],
|
| 358 |
+
[
|
| 359 |
+
1008,
|
| 360 |
+
5040
|
| 361 |
+
],
|
| 362 |
+
[
|
| 363 |
+
1008,
|
| 364 |
+
5376
|
| 365 |
+
],
|
| 366 |
+
[
|
| 367 |
+
1344,
|
| 368 |
+
336
|
| 369 |
+
],
|
| 370 |
+
[
|
| 371 |
+
1344,
|
| 372 |
+
672
|
| 373 |
+
],
|
| 374 |
+
[
|
| 375 |
+
1344,
|
| 376 |
+
1008
|
| 377 |
+
],
|
| 378 |
+
[
|
| 379 |
+
1344,
|
| 380 |
+
1344
|
| 381 |
+
],
|
| 382 |
+
[
|
| 383 |
+
1344,
|
| 384 |
+
1680
|
| 385 |
+
],
|
| 386 |
+
[
|
| 387 |
+
1344,
|
| 388 |
+
2016
|
| 389 |
+
],
|
| 390 |
+
[
|
| 391 |
+
1344,
|
| 392 |
+
2352
|
| 393 |
+
],
|
| 394 |
+
[
|
| 395 |
+
1344,
|
| 396 |
+
2688
|
| 397 |
+
],
|
| 398 |
+
[
|
| 399 |
+
1344,
|
| 400 |
+
3024
|
| 401 |
+
],
|
| 402 |
+
[
|
| 403 |
+
1344,
|
| 404 |
+
3360
|
| 405 |
+
],
|
| 406 |
+
[
|
| 407 |
+
1344,
|
| 408 |
+
3696
|
| 409 |
+
],
|
| 410 |
+
[
|
| 411 |
+
1344,
|
| 412 |
+
4032
|
| 413 |
+
],
|
| 414 |
+
[
|
| 415 |
+
1680,
|
| 416 |
+
336
|
| 417 |
+
],
|
| 418 |
+
[
|
| 419 |
+
1680,
|
| 420 |
+
672
|
| 421 |
+
],
|
| 422 |
+
[
|
| 423 |
+
1680,
|
| 424 |
+
1008
|
| 425 |
+
],
|
| 426 |
+
[
|
| 427 |
+
1680,
|
| 428 |
+
1344
|
| 429 |
+
],
|
| 430 |
+
[
|
| 431 |
+
1680,
|
| 432 |
+
1680
|
| 433 |
+
],
|
| 434 |
+
[
|
| 435 |
+
1680,
|
| 436 |
+
2016
|
| 437 |
+
],
|
| 438 |
+
[
|
| 439 |
+
1680,
|
| 440 |
+
2352
|
| 441 |
+
],
|
| 442 |
+
[
|
| 443 |
+
1680,
|
| 444 |
+
2688
|
| 445 |
+
],
|
| 446 |
+
[
|
| 447 |
+
1680,
|
| 448 |
+
3024
|
| 449 |
+
],
|
| 450 |
+
[
|
| 451 |
+
2016,
|
| 452 |
+
336
|
| 453 |
+
],
|
| 454 |
+
[
|
| 455 |
+
2016,
|
| 456 |
+
672
|
| 457 |
+
],
|
| 458 |
+
[
|
| 459 |
+
2016,
|
| 460 |
+
1008
|
| 461 |
+
],
|
| 462 |
+
[
|
| 463 |
+
2016,
|
| 464 |
+
1344
|
| 465 |
+
],
|
| 466 |
+
[
|
| 467 |
+
2016,
|
| 468 |
+
1680
|
| 469 |
+
],
|
| 470 |
+
[
|
| 471 |
+
2016,
|
| 472 |
+
2016
|
| 473 |
+
],
|
| 474 |
+
[
|
| 475 |
+
2016,
|
| 476 |
+
2352
|
| 477 |
+
],
|
| 478 |
+
[
|
| 479 |
+
2016,
|
| 480 |
+
2688
|
| 481 |
+
],
|
| 482 |
+
[
|
| 483 |
+
2352,
|
| 484 |
+
336
|
| 485 |
+
],
|
| 486 |
+
[
|
| 487 |
+
2352,
|
| 488 |
+
672
|
| 489 |
+
],
|
| 490 |
+
[
|
| 491 |
+
2352,
|
| 492 |
+
1008
|
| 493 |
+
],
|
| 494 |
+
[
|
| 495 |
+
2352,
|
| 496 |
+
1344
|
| 497 |
+
],
|
| 498 |
+
[
|
| 499 |
+
2352,
|
| 500 |
+
1680
|
| 501 |
+
],
|
| 502 |
+
[
|
| 503 |
+
2352,
|
| 504 |
+
2016
|
| 505 |
+
],
|
| 506 |
+
[
|
| 507 |
+
2352,
|
| 508 |
+
2352
|
| 509 |
+
],
|
| 510 |
+
[
|
| 511 |
+
2688,
|
| 512 |
+
336
|
| 513 |
+
],
|
| 514 |
+
[
|
| 515 |
+
2688,
|
| 516 |
+
672
|
| 517 |
+
],
|
| 518 |
+
[
|
| 519 |
+
2688,
|
| 520 |
+
1008
|
| 521 |
+
],
|
| 522 |
+
[
|
| 523 |
+
2688,
|
| 524 |
+
1344
|
| 525 |
+
],
|
| 526 |
+
[
|
| 527 |
+
2688,
|
| 528 |
+
1680
|
| 529 |
+
],
|
| 530 |
+
[
|
| 531 |
+
2688,
|
| 532 |
+
2016
|
| 533 |
+
],
|
| 534 |
+
[
|
| 535 |
+
3024,
|
| 536 |
+
336
|
| 537 |
+
],
|
| 538 |
+
[
|
| 539 |
+
3024,
|
| 540 |
+
672
|
| 541 |
+
],
|
| 542 |
+
[
|
| 543 |
+
3024,
|
| 544 |
+
1008
|
| 545 |
+
],
|
| 546 |
+
[
|
| 547 |
+
3024,
|
| 548 |
+
1344
|
| 549 |
+
],
|
| 550 |
+
[
|
| 551 |
+
3024,
|
| 552 |
+
1680
|
| 553 |
+
],
|
| 554 |
+
[
|
| 555 |
+
3360,
|
| 556 |
+
336
|
| 557 |
+
],
|
| 558 |
+
[
|
| 559 |
+
3360,
|
| 560 |
+
672
|
| 561 |
+
],
|
| 562 |
+
[
|
| 563 |
+
3360,
|
| 564 |
+
1008
|
| 565 |
+
],
|
| 566 |
+
[
|
| 567 |
+
3360,
|
| 568 |
+
1344
|
| 569 |
+
],
|
| 570 |
+
[
|
| 571 |
+
3696,
|
| 572 |
+
336
|
| 573 |
+
],
|
| 574 |
+
[
|
| 575 |
+
3696,
|
| 576 |
+
672
|
| 577 |
+
],
|
| 578 |
+
[
|
| 579 |
+
3696,
|
| 580 |
+
1008
|
| 581 |
+
],
|
| 582 |
+
[
|
| 583 |
+
3696,
|
| 584 |
+
1344
|
| 585 |
+
],
|
| 586 |
+
[
|
| 587 |
+
4032,
|
| 588 |
+
336
|
| 589 |
+
],
|
| 590 |
+
[
|
| 591 |
+
4032,
|
| 592 |
+
672
|
| 593 |
+
],
|
| 594 |
+
[
|
| 595 |
+
4032,
|
| 596 |
+
1008
|
| 597 |
+
],
|
| 598 |
+
[
|
| 599 |
+
4032,
|
| 600 |
+
1344
|
| 601 |
+
],
|
| 602 |
+
[
|
| 603 |
+
4368,
|
| 604 |
+
336
|
| 605 |
+
],
|
| 606 |
+
[
|
| 607 |
+
4368,
|
| 608 |
+
672
|
| 609 |
+
],
|
| 610 |
+
[
|
| 611 |
+
4368,
|
| 612 |
+
1008
|
| 613 |
+
],
|
| 614 |
+
[
|
| 615 |
+
4704,
|
| 616 |
+
336
|
| 617 |
+
],
|
| 618 |
+
[
|
| 619 |
+
4704,
|
| 620 |
+
672
|
| 621 |
+
],
|
| 622 |
+
[
|
| 623 |
+
4704,
|
| 624 |
+
1008
|
| 625 |
+
],
|
| 626 |
+
[
|
| 627 |
+
5040,
|
| 628 |
+
336
|
| 629 |
+
],
|
| 630 |
+
[
|
| 631 |
+
5040,
|
| 632 |
+
672
|
| 633 |
+
],
|
| 634 |
+
[
|
| 635 |
+
5040,
|
| 636 |
+
1008
|
| 637 |
+
],
|
| 638 |
+
[
|
| 639 |
+
5376,
|
| 640 |
+
336
|
| 641 |
+
],
|
| 642 |
+
[
|
| 643 |
+
5376,
|
| 644 |
+
672
|
| 645 |
+
],
|
| 646 |
+
[
|
| 647 |
+
5376,
|
| 648 |
+
1008
|
| 649 |
+
],
|
| 650 |
+
[
|
| 651 |
+
5712,
|
| 652 |
+
336
|
| 653 |
+
],
|
| 654 |
+
[
|
| 655 |
+
5712,
|
| 656 |
+
672
|
| 657 |
+
],
|
| 658 |
+
[
|
| 659 |
+
6048,
|
| 660 |
+
336
|
| 661 |
+
],
|
| 662 |
+
[
|
| 663 |
+
6048,
|
| 664 |
+
672
|
| 665 |
+
],
|
| 666 |
+
[
|
| 667 |
+
6384,
|
| 668 |
+
336
|
| 669 |
+
],
|
| 670 |
+
[
|
| 671 |
+
6384,
|
| 672 |
+
672
|
| 673 |
+
],
|
| 674 |
+
[
|
| 675 |
+
6720,
|
| 676 |
+
336
|
| 677 |
+
],
|
| 678 |
+
[
|
| 679 |
+
6720,
|
| 680 |
+
672
|
| 681 |
+
],
|
| 682 |
+
[
|
| 683 |
+
7056,
|
| 684 |
+
336
|
| 685 |
+
],
|
| 686 |
+
[
|
| 687 |
+
7056,
|
| 688 |
+
672
|
| 689 |
+
],
|
| 690 |
+
[
|
| 691 |
+
7392,
|
| 692 |
+
336
|
| 693 |
+
],
|
| 694 |
+
[
|
| 695 |
+
7392,
|
| 696 |
+
672
|
| 697 |
+
],
|
| 698 |
+
[
|
| 699 |
+
7728,
|
| 700 |
+
336
|
| 701 |
+
],
|
| 702 |
+
[
|
| 703 |
+
7728,
|
| 704 |
+
672
|
| 705 |
+
],
|
| 706 |
+
[
|
| 707 |
+
8064,
|
| 708 |
+
336
|
| 709 |
+
],
|
| 710 |
+
[
|
| 711 |
+
8064,
|
| 712 |
+
672
|
| 713 |
+
],
|
| 714 |
+
[
|
| 715 |
+
8400,
|
| 716 |
+
336
|
| 717 |
+
],
|
| 718 |
+
[
|
| 719 |
+
8736,
|
| 720 |
+
336
|
| 721 |
+
],
|
| 722 |
+
[
|
| 723 |
+
9072,
|
| 724 |
+
336
|
| 725 |
+
],
|
| 726 |
+
[
|
| 727 |
+
9408,
|
| 728 |
+
336
|
| 729 |
+
],
|
| 730 |
+
[
|
| 731 |
+
9744,
|
| 732 |
+
336
|
| 733 |
+
],
|
| 734 |
+
[
|
| 735 |
+
10080,
|
| 736 |
+
336
|
| 737 |
+
],
|
| 738 |
+
[
|
| 739 |
+
10416,
|
| 740 |
+
336
|
| 741 |
+
],
|
| 742 |
+
[
|
| 743 |
+
10752,
|
| 744 |
+
336
|
| 745 |
+
],
|
| 746 |
+
[
|
| 747 |
+
11088,
|
| 748 |
+
336
|
| 749 |
+
],
|
| 750 |
+
[
|
| 751 |
+
11424,
|
| 752 |
+
336
|
| 753 |
+
],
|
| 754 |
+
[
|
| 755 |
+
11760,
|
| 756 |
+
336
|
| 757 |
+
],
|
| 758 |
+
[
|
| 759 |
+
12096,
|
| 760 |
+
336
|
| 761 |
+
],
|
| 762 |
+
[
|
| 763 |
+
12432,
|
| 764 |
+
336
|
| 765 |
+
],
|
| 766 |
+
[
|
| 767 |
+
12768,
|
| 768 |
+
336
|
| 769 |
+
],
|
| 770 |
+
[
|
| 771 |
+
13104,
|
| 772 |
+
336
|
| 773 |
+
],
|
| 774 |
+
[
|
| 775 |
+
13440,
|
| 776 |
+
336
|
| 777 |
+
],
|
| 778 |
+
[
|
| 779 |
+
13776,
|
| 780 |
+
336
|
| 781 |
+
],
|
| 782 |
+
[
|
| 783 |
+
14112,
|
| 784 |
+
336
|
| 785 |
+
],
|
| 786 |
+
[
|
| 787 |
+
14448,
|
| 788 |
+
336
|
| 789 |
+
],
|
| 790 |
+
[
|
| 791 |
+
14784,
|
| 792 |
+
336
|
| 793 |
+
],
|
| 794 |
+
[
|
| 795 |
+
15120,
|
| 796 |
+
336
|
| 797 |
+
],
|
| 798 |
+
[
|
| 799 |
+
15456,
|
| 800 |
+
336
|
| 801 |
+
],
|
| 802 |
+
[
|
| 803 |
+
15792,
|
| 804 |
+
336
|
| 805 |
+
],
|
| 806 |
+
[
|
| 807 |
+
16128,
|
| 808 |
+
336
|
| 809 |
+
],
|
| 810 |
+
[
|
| 811 |
+
16464,
|
| 812 |
+
336
|
| 813 |
+
]
|
| 814 |
+
],
|
| 815 |
+
"image_split_resolution": null,
|
| 816 |
+
"initializer_range": 0.02,
|
| 817 |
+
"intermediate_size": 18944,
|
| 818 |
+
"max_position_embeddings": 224000,
|
| 819 |
+
"max_window_layers": 28,
|
| 820 |
+
"mm_hidden_size": 1024,
|
| 821 |
+
"mm_patch_merge_type": "unires",
|
| 822 |
+
"mm_projector_lr": null,
|
| 823 |
+
"mm_projector_type": "mlp2x_gelu",
|
| 824 |
+
"mm_resampler_type": null,
|
| 825 |
+
"mm_spatial_pool_mode": "average",
|
| 826 |
+
"mm_spatial_pool_stride": 2,
|
| 827 |
+
"mm_tunable_parts": "speech_projector,mm_mlp_adapter,mm_language_model",
|
| 828 |
+
"mm_use_im_patch_token": false,
|
| 829 |
+
"mm_use_im_start_end": false,
|
| 830 |
+
"mm_vision_select_feature": "patch",
|
| 831 |
+
"mm_vision_select_layer": -2,
|
| 832 |
+
"mm_vision_tower": "checkpoints/clip-vit-large-patch14-336",
|
| 833 |
+
"mm_vision_tower_lr": 2e-06,
|
| 834 |
+
"model_type": "qwen2",
|
| 835 |
+
"num_attention_heads": 28,
|
| 836 |
+
"num_hidden_layers": 28,
|
| 837 |
+
"num_key_value_heads": 4,
|
| 838 |
+
"pos_skipping_range": 4096,
|
| 839 |
+
"rms_norm_eps": 1e-06,
|
| 840 |
+
"rope_scaling": null,
|
| 841 |
+
"rope_theta": 1000000000.0,
|
| 842 |
+
"sliding_window": null,
|
| 843 |
+
"speech_encoder": "checkpoints/whisper/large-v3.pt",
|
| 844 |
+
"speech_encoder_ds_rate": 5,
|
| 845 |
+
"speech_encoder_hidden_size": 1280,
|
| 846 |
+
"speech_encoder_type": "whisper",
|
| 847 |
+
"speech_normalize": false,
|
| 848 |
+
"speech_projector_lr": null,
|
| 849 |
+
"speech_projector_type": "linear",
|
| 850 |
+
"tie_word_embeddings": false,
|
| 851 |
+
"tokenizer_model_max_length": 16384,
|
| 852 |
+
"tokenizer_padding_side": "right",
|
| 853 |
+
"torch_dtype": "bfloat16",
|
| 854 |
+
"transformers_version": "4.44.0",
|
| 855 |
+
"use_cache": true,
|
| 856 |
+
"use_mm_proj": true,
|
| 857 |
+
"use_pos_skipping": false,
|
| 858 |
+
"use_sliding_window": false,
|
| 859 |
+
"vision_tower_pretrained": null,
|
| 860 |
+
"vocab_size": 152064
|
| 861 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_attn_implementation": "flash_attention_2",
|
| 3 |
+
"bos_token_id": 151643,
|
| 4 |
+
"do_sample": true,
|
| 5 |
+
"eos_token_id": [
|
| 6 |
+
151645,
|
| 7 |
+
151643
|
| 8 |
+
],
|
| 9 |
+
"pad_token_id": 151643,
|
| 10 |
+
"repetition_penalty": 1.05,
|
| 11 |
+
"rope_theta": 1000000000.0,
|
| 12 |
+
"temperature": 0.7,
|
| 13 |
+
"top_k": 20,
|
| 14 |
+
"top_p": 0.8,
|
| 15 |
+
"transformers_version": "4.44.0"
|
| 16 |
+
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
model-00001-of-00004.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:eba4d40cf97b4f248a4792134487227c9d7039a238e16eab992a594e9c27d64e
|
| 3 |
+
size 4877660776
|
model-00002-of-00004.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e7f8beaca936c534a2389ac8189ed0d43cc25229f669bfc0eeeec7915f587fa0
|
| 3 |
+
size 4932751008
|
model-00003-of-00004.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:03ea0b848fe54b19ec312e9782030f627c5a6d5a09d68aad294e555903a5f32c
|
| 3 |
+
size 4998766592
|
model-00004-of-00004.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:768f0129d73f42fdb45c423258b77a0ed5d1b4ae296d689ae4300f19f914f6a2
|
| 3 |
+
size 2377117192
|
model.safetensors.index.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<|im_start|>",
|
| 4 |
+
"<|im_end|>"
|
| 5 |
+
],
|
| 6 |
+
"eos_token": {
|
| 7 |
+
"content": "<|im_end|>",
|
| 8 |
+
"lstrip": false,
|
| 9 |
+
"normalized": false,
|
| 10 |
+
"rstrip": false,
|
| 11 |
+
"single_word": false
|
| 12 |
+
},
|
| 13 |
+
"pad_token": {
|
| 14 |
+
"content": "<|endoftext|>",
|
| 15 |
+
"lstrip": false,
|
| 16 |
+
"normalized": false,
|
| 17 |
+
"rstrip": false,
|
| 18 |
+
"single_word": false
|
| 19 |
+
}
|
| 20 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"151643": {
|
| 5 |
+
"content": "<|endoftext|>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": false,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"151644": {
|
| 13 |
+
"content": "<|im_start|>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": false,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"151645": {
|
| 21 |
+
"content": "<|im_end|>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": false,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": true
|
| 27 |
+
}
|
| 28 |
+
},
|
| 29 |
+
"additional_special_tokens": [
|
| 30 |
+
"<|im_start|>",
|
| 31 |
+
"<|im_end|>"
|
| 32 |
+
],
|
| 33 |
+
"bos_token": null,
|
| 34 |
+
"chat_template": "{% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n' }}{% endif %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
|
| 35 |
+
"clean_up_tokenization_spaces": false,
|
| 36 |
+
"eos_token": "<|im_end|>",
|
| 37 |
+
"errors": "replace",
|
| 38 |
+
"model_max_length": 16384,
|
| 39 |
+
"pad_token": "<|endoftext|>",
|
| 40 |
+
"padding_side": "right",
|
| 41 |
+
"split_special_tokens": false,
|
| 42 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 43 |
+
"unk_token": null
|
| 44 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2219 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9991967871485944,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 311,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.00321285140562249,
|
| 13 |
+
"grad_norm": 9.468059539794922,
|
| 14 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 15 |
+
"loss": 1.1591,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.00642570281124498,
|
| 20 |
+
"grad_norm": 12.971900939941406,
|
| 21 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 22 |
+
"loss": 0.9396,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.00963855421686747,
|
| 27 |
+
"grad_norm": 9.142277717590332,
|
| 28 |
+
"learning_rate": 3e-06,
|
| 29 |
+
"loss": 1.0339,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.01285140562248996,
|
| 34 |
+
"grad_norm": 7.4195380210876465,
|
| 35 |
+
"learning_rate": 4.000000000000001e-06,
|
| 36 |
+
"loss": 1.0046,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.01606425702811245,
|
| 41 |
+
"grad_norm": 7.382334232330322,
|
| 42 |
+
"learning_rate": 5e-06,
|
| 43 |
+
"loss": 0.8551,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.01927710843373494,
|
| 48 |
+
"grad_norm": 5.827909469604492,
|
| 49 |
+
"learning_rate": 6e-06,
|
| 50 |
+
"loss": 0.8208,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.02248995983935743,
|
| 55 |
+
"grad_norm": 4.3395795822143555,
|
| 56 |
+
"learning_rate": 7e-06,
|
| 57 |
+
"loss": 0.6632,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.02570281124497992,
|
| 62 |
+
"grad_norm": 34.95582962036133,
|
| 63 |
+
"learning_rate": 8.000000000000001e-06,
|
| 64 |
+
"loss": 1.1221,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.02891566265060241,
|
| 69 |
+
"grad_norm": 5.787684440612793,
|
| 70 |
+
"learning_rate": 9e-06,
|
| 71 |
+
"loss": 0.7092,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.0321285140562249,
|
| 76 |
+
"grad_norm": 3.8266000747680664,
|
| 77 |
+
"learning_rate": 1e-05,
|
| 78 |
+
"loss": 0.8341,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.035341365461847386,
|
| 83 |
+
"grad_norm": 7.160825252532959,
|
| 84 |
+
"learning_rate": 9.999727665400876e-06,
|
| 85 |
+
"loss": 0.7178,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.03855421686746988,
|
| 90 |
+
"grad_norm": 3.6943607330322266,
|
| 91 |
+
"learning_rate": 9.998910691269957e-06,
|
| 92 |
+
"loss": 0.8316,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.04176706827309237,
|
| 97 |
+
"grad_norm": 4.6069159507751465,
|
| 98 |
+
"learning_rate": 9.99754916660337e-06,
|
| 99 |
+
"loss": 0.631,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.04497991967871486,
|
| 104 |
+
"grad_norm": 4.735833168029785,
|
| 105 |
+
"learning_rate": 9.995643239717228e-06,
|
| 106 |
+
"loss": 0.7483,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.04819277108433735,
|
| 111 |
+
"grad_norm": 3.0799083709716797,
|
| 112 |
+
"learning_rate": 9.993193118231463e-06,
|
| 113 |
+
"loss": 0.4765,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.05140562248995984,
|
| 118 |
+
"grad_norm": 2.8460216522216797,
|
| 119 |
+
"learning_rate": 9.990199069047216e-06,
|
| 120 |
+
"loss": 0.6183,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.05461847389558233,
|
| 125 |
+
"grad_norm": 3.5829732418060303,
|
| 126 |
+
"learning_rate": 9.986661418317759e-06,
|
| 127 |
+
"loss": 0.7606,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.05783132530120482,
|
| 132 |
+
"grad_norm": 2.786634683609009,
|
| 133 |
+
"learning_rate": 9.982580551412972e-06,
|
| 134 |
+
"loss": 0.6369,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.06104417670682731,
|
| 139 |
+
"grad_norm": 3.3990700244903564,
|
| 140 |
+
"learning_rate": 9.977956912877356e-06,
|
| 141 |
+
"loss": 0.5388,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.0642570281124498,
|
| 146 |
+
"grad_norm": 2.86079478263855,
|
| 147 |
+
"learning_rate": 9.97279100638161e-06,
|
| 148 |
+
"loss": 0.7238,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.06746987951807229,
|
| 153 |
+
"grad_norm": 2.6607353687286377,
|
| 154 |
+
"learning_rate": 9.967083394667763e-06,
|
| 155 |
+
"loss": 0.6505,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.07068273092369477,
|
| 160 |
+
"grad_norm": 2.682445526123047,
|
| 161 |
+
"learning_rate": 9.960834699487873e-06,
|
| 162 |
+
"loss": 0.4418,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.07389558232931727,
|
| 167 |
+
"grad_norm": 2.598595380783081,
|
| 168 |
+
"learning_rate": 9.9540456015363e-06,
|
| 169 |
+
"loss": 0.616,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.07710843373493977,
|
| 174 |
+
"grad_norm": 3.148313045501709,
|
| 175 |
+
"learning_rate": 9.946716840375552e-06,
|
| 176 |
+
"loss": 0.5655,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.08032128514056225,
|
| 181 |
+
"grad_norm": 3.6518056392669678,
|
| 182 |
+
"learning_rate": 9.938849214355722e-06,
|
| 183 |
+
"loss": 0.5541,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.08353413654618475,
|
| 188 |
+
"grad_norm": 2.7225069999694824,
|
| 189 |
+
"learning_rate": 9.93044358052752e-06,
|
| 190 |
+
"loss": 0.6182,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.08674698795180723,
|
| 195 |
+
"grad_norm": 2.9630682468414307,
|
| 196 |
+
"learning_rate": 9.921500854548916e-06,
|
| 197 |
+
"loss": 0.5794,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.08995983935742972,
|
| 202 |
+
"grad_norm": 2.659717082977295,
|
| 203 |
+
"learning_rate": 9.912022010585385e-06,
|
| 204 |
+
"loss": 0.5407,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.09317269076305221,
|
| 209 |
+
"grad_norm": 2.64546799659729,
|
| 210 |
+
"learning_rate": 9.902008081203796e-06,
|
| 211 |
+
"loss": 0.5833,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.0963855421686747,
|
| 216 |
+
"grad_norm": 3.679818868637085,
|
| 217 |
+
"learning_rate": 9.89146015725993e-06,
|
| 218 |
+
"loss": 0.6286,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.09959839357429719,
|
| 223 |
+
"grad_norm": 3.6220390796661377,
|
| 224 |
+
"learning_rate": 9.880379387779637e-06,
|
| 225 |
+
"loss": 0.717,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.10281124497991968,
|
| 230 |
+
"grad_norm": 2.7515664100646973,
|
| 231 |
+
"learning_rate": 9.868766979833686e-06,
|
| 232 |
+
"loss": 0.5431,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.10602409638554217,
|
| 237 |
+
"grad_norm": 2.658803939819336,
|
| 238 |
+
"learning_rate": 9.856624198406262e-06,
|
| 239 |
+
"loss": 0.6401,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.10923694779116466,
|
| 244 |
+
"grad_norm": 3.2038729190826416,
|
| 245 |
+
"learning_rate": 9.84395236625717e-06,
|
| 246 |
+
"loss": 0.5142,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.11244979919678715,
|
| 251 |
+
"grad_norm": 4.682802200317383,
|
| 252 |
+
"learning_rate": 9.830752863777741e-06,
|
| 253 |
+
"loss": 0.6152,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.11566265060240964,
|
| 258 |
+
"grad_norm": 2.9723594188690186,
|
| 259 |
+
"learning_rate": 9.817027128840462e-06,
|
| 260 |
+
"loss": 0.3418,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.11887550200803212,
|
| 265 |
+
"grad_norm": 3.891775369644165,
|
| 266 |
+
"learning_rate": 9.802776656642341e-06,
|
| 267 |
+
"loss": 0.6575,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.12208835341365462,
|
| 272 |
+
"grad_norm": 2.52054762840271,
|
| 273 |
+
"learning_rate": 9.78800299954203e-06,
|
| 274 |
+
"loss": 0.5537,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.12530120481927712,
|
| 279 |
+
"grad_norm": 2.726982593536377,
|
| 280 |
+
"learning_rate": 9.772707766890726e-06,
|
| 281 |
+
"loss": 0.6654,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.1285140562248996,
|
| 286 |
+
"grad_norm": 2.5964579582214355,
|
| 287 |
+
"learning_rate": 9.756892624856848e-06,
|
| 288 |
+
"loss": 0.5041,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.13172690763052208,
|
| 293 |
+
"grad_norm": 3.1076455116271973,
|
| 294 |
+
"learning_rate": 9.740559296244543e-06,
|
| 295 |
+
"loss": 0.4663,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.13493975903614458,
|
| 300 |
+
"grad_norm": 3.2797796726226807,
|
| 301 |
+
"learning_rate": 9.723709560306009e-06,
|
| 302 |
+
"loss": 0.5047,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.13815261044176708,
|
| 307 |
+
"grad_norm": 2.475914478302002,
|
| 308 |
+
"learning_rate": 9.706345252547681e-06,
|
| 309 |
+
"loss": 0.6025,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.14136546184738955,
|
| 314 |
+
"grad_norm": 2.9697823524475098,
|
| 315 |
+
"learning_rate": 9.688468264530278e-06,
|
| 316 |
+
"loss": 0.3588,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.14457831325301204,
|
| 321 |
+
"grad_norm": 2.9819068908691406,
|
| 322 |
+
"learning_rate": 9.670080543662742e-06,
|
| 323 |
+
"loss": 0.5415,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.14779116465863454,
|
| 328 |
+
"grad_norm": 2.63847017288208,
|
| 329 |
+
"learning_rate": 9.651184092990109e-06,
|
| 330 |
+
"loss": 0.548,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.15100401606425704,
|
| 335 |
+
"grad_norm": 2.9994335174560547,
|
| 336 |
+
"learning_rate": 9.631780970975311e-06,
|
| 337 |
+
"loss": 0.5151,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.15421686746987953,
|
| 342 |
+
"grad_norm": 2.440392017364502,
|
| 343 |
+
"learning_rate": 9.611873291274927e-06,
|
| 344 |
+
"loss": 0.4567,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.157429718875502,
|
| 349 |
+
"grad_norm": 2.510368585586548,
|
| 350 |
+
"learning_rate": 9.591463222508947e-06,
|
| 351 |
+
"loss": 0.5311,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.1606425702811245,
|
| 356 |
+
"grad_norm": 2.9306483268737793,
|
| 357 |
+
"learning_rate": 9.570552988024527e-06,
|
| 358 |
+
"loss": 0.3788,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.163855421686747,
|
| 363 |
+
"grad_norm": 3.0692126750946045,
|
| 364 |
+
"learning_rate": 9.5491448656538e-06,
|
| 365 |
+
"loss": 0.5438,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.1670682730923695,
|
| 370 |
+
"grad_norm": 2.9109346866607666,
|
| 371 |
+
"learning_rate": 9.527241187465735e-06,
|
| 372 |
+
"loss": 0.5757,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.17028112449799196,
|
| 377 |
+
"grad_norm": 2.6207594871520996,
|
| 378 |
+
"learning_rate": 9.504844339512096e-06,
|
| 379 |
+
"loss": 0.5567,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.17349397590361446,
|
| 384 |
+
"grad_norm": 2.2211403846740723,
|
| 385 |
+
"learning_rate": 9.481956761567531e-06,
|
| 386 |
+
"loss": 0.3586,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.17670682730923695,
|
| 391 |
+
"grad_norm": 2.620298147201538,
|
| 392 |
+
"learning_rate": 9.458580946863784e-06,
|
| 393 |
+
"loss": 0.4068,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.17991967871485945,
|
| 398 |
+
"grad_norm": 3.2399895191192627,
|
| 399 |
+
"learning_rate": 9.434719441818106e-06,
|
| 400 |
+
"loss": 0.4569,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.18313253012048192,
|
| 405 |
+
"grad_norm": 3.0689821243286133,
|
| 406 |
+
"learning_rate": 9.410374845755862e-06,
|
| 407 |
+
"loss": 0.4875,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.18634538152610441,
|
| 412 |
+
"grad_norm": 2.3414645195007324,
|
| 413 |
+
"learning_rate": 9.385549810627374e-06,
|
| 414 |
+
"loss": 0.4211,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.1895582329317269,
|
| 419 |
+
"grad_norm": 2.688450336456299,
|
| 420 |
+
"learning_rate": 9.36024704071904e-06,
|
| 421 |
+
"loss": 0.3738,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.1927710843373494,
|
| 426 |
+
"grad_norm": 2.5549943447113037,
|
| 427 |
+
"learning_rate": 9.334469292358736e-06,
|
| 428 |
+
"loss": 0.4483,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.19598393574297188,
|
| 433 |
+
"grad_norm": 3.2717857360839844,
|
| 434 |
+
"learning_rate": 9.308219373615574e-06,
|
| 435 |
+
"loss": 0.4543,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.19919678714859437,
|
| 440 |
+
"grad_norm": 2.5511434078216553,
|
| 441 |
+
"learning_rate": 9.28150014399399e-06,
|
| 442 |
+
"loss": 0.5196,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.20240963855421687,
|
| 447 |
+
"grad_norm": 4.289268970489502,
|
| 448 |
+
"learning_rate": 9.25431451412226e-06,
|
| 449 |
+
"loss": 0.4566,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.20562248995983937,
|
| 454 |
+
"grad_norm": 2.4556353092193604,
|
| 455 |
+
"learning_rate": 9.226665445435428e-06,
|
| 456 |
+
"loss": 0.4854,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.20883534136546184,
|
| 461 |
+
"grad_norm": 2.211134910583496,
|
| 462 |
+
"learning_rate": 9.19855594985271e-06,
|
| 463 |
+
"loss": 0.4626,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.21204819277108433,
|
| 468 |
+
"grad_norm": 2.233396530151367,
|
| 469 |
+
"learning_rate": 9.16998908944939e-06,
|
| 470 |
+
"loss": 0.4688,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.21526104417670683,
|
| 475 |
+
"grad_norm": 2.952958106994629,
|
| 476 |
+
"learning_rate": 9.14096797612326e-06,
|
| 477 |
+
"loss": 0.5239,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.21847389558232932,
|
| 482 |
+
"grad_norm": 2.30307936668396,
|
| 483 |
+
"learning_rate": 9.111495771255623e-06,
|
| 484 |
+
"loss": 0.4765,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.2216867469879518,
|
| 489 |
+
"grad_norm": 3.2021255493164062,
|
| 490 |
+
"learning_rate": 9.081575685366919e-06,
|
| 491 |
+
"loss": 0.579,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.2248995983935743,
|
| 496 |
+
"grad_norm": 2.920560359954834,
|
| 497 |
+
"learning_rate": 9.051210977766987e-06,
|
| 498 |
+
"loss": 0.4866,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.2281124497991968,
|
| 503 |
+
"grad_norm": 2.718421697616577,
|
| 504 |
+
"learning_rate": 9.020404956200016e-06,
|
| 505 |
+
"loss": 0.4484,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.23132530120481928,
|
| 510 |
+
"grad_norm": 2.929945707321167,
|
| 511 |
+
"learning_rate": 8.989160976484218e-06,
|
| 512 |
+
"loss": 0.4432,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.23453815261044178,
|
| 517 |
+
"grad_norm": 3.0811638832092285,
|
| 518 |
+
"learning_rate": 8.957482442146271e-06,
|
| 519 |
+
"loss": 0.5614,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.23775100401606425,
|
| 524 |
+
"grad_norm": 3.6012563705444336,
|
| 525 |
+
"learning_rate": 8.925372804050554e-06,
|
| 526 |
+
"loss": 0.4146,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.24096385542168675,
|
| 531 |
+
"grad_norm": 2.4589638710021973,
|
| 532 |
+
"learning_rate": 8.892835560023236e-06,
|
| 533 |
+
"loss": 0.4544,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.24417670682730924,
|
| 538 |
+
"grad_norm": 2.542167901992798,
|
| 539 |
+
"learning_rate": 8.85987425447124e-06,
|
| 540 |
+
"loss": 0.5399,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.24738955823293174,
|
| 545 |
+
"grad_norm": 2.2393405437469482,
|
| 546 |
+
"learning_rate": 8.826492477996138e-06,
|
| 547 |
+
"loss": 0.4348,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.25060240963855424,
|
| 552 |
+
"grad_norm": 2.5719082355499268,
|
| 553 |
+
"learning_rate": 8.792693867003017e-06,
|
| 554 |
+
"loss": 0.5533,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.25381526104417673,
|
| 559 |
+
"grad_norm": 2.2429287433624268,
|
| 560 |
+
"learning_rate": 8.758482103304348e-06,
|
| 561 |
+
"loss": 0.5089,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.2570281124497992,
|
| 566 |
+
"grad_norm": 2.215193033218384,
|
| 567 |
+
"learning_rate": 8.72386091371891e-06,
|
| 568 |
+
"loss": 0.3864,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.26024096385542167,
|
| 573 |
+
"grad_norm": 2.946692943572998,
|
| 574 |
+
"learning_rate": 8.688834069665819e-06,
|
| 575 |
+
"loss": 0.4929,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.26345381526104417,
|
| 580 |
+
"grad_norm": 2.1577095985412598,
|
| 581 |
+
"learning_rate": 8.653405386753688e-06,
|
| 582 |
+
"loss": 0.3006,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.26666666666666666,
|
| 587 |
+
"grad_norm": 2.4236855506896973,
|
| 588 |
+
"learning_rate": 8.617578724364984e-06,
|
| 589 |
+
"loss": 0.3955,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.26987951807228916,
|
| 594 |
+
"grad_norm": 2.1099393367767334,
|
| 595 |
+
"learning_rate": 8.581357985235595e-06,
|
| 596 |
+
"loss": 0.289,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.27309236947791166,
|
| 601 |
+
"grad_norm": 2.321336030960083,
|
| 602 |
+
"learning_rate": 8.544747115029717e-06,
|
| 603 |
+
"loss": 0.3508,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.27630522088353415,
|
| 608 |
+
"grad_norm": 3.0077357292175293,
|
| 609 |
+
"learning_rate": 8.50775010191001e-06,
|
| 610 |
+
"loss": 0.38,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.27951807228915665,
|
| 615 |
+
"grad_norm": 2.451876640319824,
|
| 616 |
+
"learning_rate": 8.470370976103171e-06,
|
| 617 |
+
"loss": 0.3444,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.2827309236947791,
|
| 622 |
+
"grad_norm": 2.5099685192108154,
|
| 623 |
+
"learning_rate": 8.432613809460895e-06,
|
| 624 |
+
"loss": 0.5357,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.2859437751004016,
|
| 629 |
+
"grad_norm": 2.212970018386841,
|
| 630 |
+
"learning_rate": 8.394482715016318e-06,
|
| 631 |
+
"loss": 0.361,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.2891566265060241,
|
| 636 |
+
"grad_norm": 2.779324531555176,
|
| 637 |
+
"learning_rate": 8.355981846535972e-06,
|
| 638 |
+
"loss": 0.5175,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.2923694779116466,
|
| 643 |
+
"grad_norm": 2.486966133117676,
|
| 644 |
+
"learning_rate": 8.317115398067289e-06,
|
| 645 |
+
"loss": 0.4769,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.2955823293172691,
|
| 650 |
+
"grad_norm": 2.3630478382110596,
|
| 651 |
+
"learning_rate": 8.27788760348173e-06,
|
| 652 |
+
"loss": 0.3273,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.2987951807228916,
|
| 657 |
+
"grad_norm": 3.0468597412109375,
|
| 658 |
+
"learning_rate": 8.238302736013587e-06,
|
| 659 |
+
"loss": 0.465,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.30200803212851407,
|
| 664 |
+
"grad_norm": 2.1398801803588867,
|
| 665 |
+
"learning_rate": 8.198365107794457e-06,
|
| 666 |
+
"loss": 0.5196,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.30522088353413657,
|
| 671 |
+
"grad_norm": 2.3869407176971436,
|
| 672 |
+
"learning_rate": 8.158079069383535e-06,
|
| 673 |
+
"loss": 0.3714,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.30843373493975906,
|
| 678 |
+
"grad_norm": 2.6658105850219727,
|
| 679 |
+
"learning_rate": 8.117449009293668e-06,
|
| 680 |
+
"loss": 0.4704,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.3116465863453815,
|
| 685 |
+
"grad_norm": 3.014535665512085,
|
| 686 |
+
"learning_rate": 8.076479353513308e-06,
|
| 687 |
+
"loss": 0.4581,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.314859437751004,
|
| 692 |
+
"grad_norm": 2.575618028640747,
|
| 693 |
+
"learning_rate": 8.035174565024362e-06,
|
| 694 |
+
"loss": 0.4222,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.3180722891566265,
|
| 699 |
+
"grad_norm": 2.499056577682495,
|
| 700 |
+
"learning_rate": 7.993539143316044e-06,
|
| 701 |
+
"loss": 0.3373,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.321285140562249,
|
| 706 |
+
"grad_norm": 1.9054265022277832,
|
| 707 |
+
"learning_rate": 7.951577623894701e-06,
|
| 708 |
+
"loss": 0.4234,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.3244979919678715,
|
| 713 |
+
"grad_norm": 2.496711015701294,
|
| 714 |
+
"learning_rate": 7.909294577789765e-06,
|
| 715 |
+
"loss": 0.3642,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.327710843373494,
|
| 720 |
+
"grad_norm": 2.4711644649505615,
|
| 721 |
+
"learning_rate": 7.866694611055796e-06,
|
| 722 |
+
"loss": 0.3368,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.3309236947791165,
|
| 727 |
+
"grad_norm": 2.246197462081909,
|
| 728 |
+
"learning_rate": 7.823782364270743e-06,
|
| 729 |
+
"loss": 0.3013,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.334136546184739,
|
| 734 |
+
"grad_norm": 2.3415684700012207,
|
| 735 |
+
"learning_rate": 7.780562512030414e-06,
|
| 736 |
+
"loss": 0.4313,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.3373493975903614,
|
| 741 |
+
"grad_norm": 2.7522289752960205,
|
| 742 |
+
"learning_rate": 7.737039762439263e-06,
|
| 743 |
+
"loss": 0.5178,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.3405622489959839,
|
| 748 |
+
"grad_norm": 2.612412452697754,
|
| 749 |
+
"learning_rate": 7.693218856597515e-06,
|
| 750 |
+
"loss": 0.4683,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.3437751004016064,
|
| 755 |
+
"grad_norm": 2.491758108139038,
|
| 756 |
+
"learning_rate": 7.649104568084701e-06,
|
| 757 |
+
"loss": 0.4714,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.3469879518072289,
|
| 762 |
+
"grad_norm": 2.803365707397461,
|
| 763 |
+
"learning_rate": 7.604701702439652e-06,
|
| 764 |
+
"loss": 0.4583,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.3502008032128514,
|
| 769 |
+
"grad_norm": 2.161888360977173,
|
| 770 |
+
"learning_rate": 7.560015096637015e-06,
|
| 771 |
+
"loss": 0.3207,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.3534136546184739,
|
| 776 |
+
"grad_norm": 2.9406943321228027,
|
| 777 |
+
"learning_rate": 7.515049618560337e-06,
|
| 778 |
+
"loss": 0.4199,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.3566265060240964,
|
| 783 |
+
"grad_norm": 2.4751358032226562,
|
| 784 |
+
"learning_rate": 7.469810166471802e-06,
|
| 785 |
+
"loss": 0.3597,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.3598393574297189,
|
| 790 |
+
"grad_norm": 2.4176955223083496,
|
| 791 |
+
"learning_rate": 7.424301668478626e-06,
|
| 792 |
+
"loss": 0.4643,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.36305220883534134,
|
| 797 |
+
"grad_norm": 1.987687110900879,
|
| 798 |
+
"learning_rate": 7.378529081996233e-06,
|
| 799 |
+
"loss": 0.3283,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.36626506024096384,
|
| 804 |
+
"grad_norm": 2.1835029125213623,
|
| 805 |
+
"learning_rate": 7.332497393208221e-06,
|
| 806 |
+
"loss": 0.3521,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.36947791164658633,
|
| 811 |
+
"grad_norm": 2.5302932262420654,
|
| 812 |
+
"learning_rate": 7.286211616523193e-06,
|
| 813 |
+
"loss": 0.397,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.37269076305220883,
|
| 818 |
+
"grad_norm": 2.3871219158172607,
|
| 819 |
+
"learning_rate": 7.239676794028526e-06,
|
| 820 |
+
"loss": 0.4685,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.3759036144578313,
|
| 825 |
+
"grad_norm": 2.5505502223968506,
|
| 826 |
+
"learning_rate": 7.192897994941111e-06,
|
| 827 |
+
"loss": 0.4473,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.3791164658634538,
|
| 832 |
+
"grad_norm": 2.5716140270233154,
|
| 833 |
+
"learning_rate": 7.145880315055145e-06,
|
| 834 |
+
"loss": 0.2596,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.3823293172690763,
|
| 839 |
+
"grad_norm": 2.9478588104248047,
|
| 840 |
+
"learning_rate": 7.098628876187031e-06,
|
| 841 |
+
"loss": 0.3903,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.3855421686746988,
|
| 846 |
+
"grad_norm": 2.7252838611602783,
|
| 847 |
+
"learning_rate": 7.051148825617435e-06,
|
| 848 |
+
"loss": 0.3011,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.3887550200803213,
|
| 853 |
+
"grad_norm": 1.9095062017440796,
|
| 854 |
+
"learning_rate": 7.003445335530572e-06,
|
| 855 |
+
"loss": 0.242,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.39196787148594375,
|
| 860 |
+
"grad_norm": 2.5876998901367188,
|
| 861 |
+
"learning_rate": 6.95552360245078e-06,
|
| 862 |
+
"loss": 0.4669,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.39518072289156625,
|
| 867 |
+
"grad_norm": 2.438539505004883,
|
| 868 |
+
"learning_rate": 6.9073888466764495e-06,
|
| 869 |
+
"loss": 0.2939,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.39839357429718875,
|
| 874 |
+
"grad_norm": 2.326950788497925,
|
| 875 |
+
"learning_rate": 6.859046311711344e-06,
|
| 876 |
+
"loss": 0.3136,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.40160642570281124,
|
| 881 |
+
"grad_norm": 2.442441463470459,
|
| 882 |
+
"learning_rate": 6.810501263693416e-06,
|
| 883 |
+
"loss": 0.241,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.40481927710843374,
|
| 888 |
+
"grad_norm": 2.1230807304382324,
|
| 889 |
+
"learning_rate": 6.761758990821143e-06,
|
| 890 |
+
"loss": 0.341,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.40803212851405624,
|
| 895 |
+
"grad_norm": 2.29844069480896,
|
| 896 |
+
"learning_rate": 6.712824802777465e-06,
|
| 897 |
+
"loss": 0.2936,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.41124497991967873,
|
| 902 |
+
"grad_norm": 2.681148052215576,
|
| 903 |
+
"learning_rate": 6.66370403015137e-06,
|
| 904 |
+
"loss": 0.271,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.41445783132530123,
|
| 909 |
+
"grad_norm": 2.411924123764038,
|
| 910 |
+
"learning_rate": 6.614402023857231e-06,
|
| 911 |
+
"loss": 0.4166,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.41767068273092367,
|
| 916 |
+
"grad_norm": 2.795295238494873,
|
| 917 |
+
"learning_rate": 6.564924154551895e-06,
|
| 918 |
+
"loss": 0.3683,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.42088353413654617,
|
| 923 |
+
"grad_norm": 2.0868473052978516,
|
| 924 |
+
"learning_rate": 6.515275812049644e-06,
|
| 925 |
+
"loss": 0.2563,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.42409638554216866,
|
| 930 |
+
"grad_norm": 2.622237205505371,
|
| 931 |
+
"learning_rate": 6.4654624047350575e-06,
|
| 932 |
+
"loss": 0.3725,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.42730923694779116,
|
| 937 |
+
"grad_norm": 2.958596706390381,
|
| 938 |
+
"learning_rate": 6.41548935897386e-06,
|
| 939 |
+
"loss": 0.2818,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.43052208835341366,
|
| 944 |
+
"grad_norm": 2.81485915184021,
|
| 945 |
+
"learning_rate": 6.365362118521807e-06,
|
| 946 |
+
"loss": 0.3887,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.43373493975903615,
|
| 951 |
+
"grad_norm": 2.5471572875976562,
|
| 952 |
+
"learning_rate": 6.31508614393167e-06,
|
| 953 |
+
"loss": 0.3455,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.43694779116465865,
|
| 958 |
+
"grad_norm": 5.213260173797607,
|
| 959 |
+
"learning_rate": 6.264666911958404e-06,
|
| 960 |
+
"loss": 0.3154,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.44016064257028115,
|
| 965 |
+
"grad_norm": 2.910991668701172,
|
| 966 |
+
"learning_rate": 6.214109914962542e-06,
|
| 967 |
+
"loss": 0.3963,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.4433734939759036,
|
| 972 |
+
"grad_norm": 2.244249105453491,
|
| 973 |
+
"learning_rate": 6.1634206603118844e-06,
|
| 974 |
+
"loss": 0.2974,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.4465863453815261,
|
| 979 |
+
"grad_norm": 2.3065643310546875,
|
| 980 |
+
"learning_rate": 6.112604669781572e-06,
|
| 981 |
+
"loss": 0.344,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.4497991967871486,
|
| 986 |
+
"grad_norm": 2.094327211380005,
|
| 987 |
+
"learning_rate": 6.06166747895257e-06,
|
| 988 |
+
"loss": 0.3327,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.4530120481927711,
|
| 993 |
+
"grad_norm": 2.5069355964660645,
|
| 994 |
+
"learning_rate": 6.0106146366086514e-06,
|
| 995 |
+
"loss": 0.2542,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.4562248995983936,
|
| 1000 |
+
"grad_norm": 2.227370262145996,
|
| 1001 |
+
"learning_rate": 5.959451704131962e-06,
|
| 1002 |
+
"loss": 0.3063,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.45943775100401607,
|
| 1007 |
+
"grad_norm": 3.291332244873047,
|
| 1008 |
+
"learning_rate": 5.908184254897183e-06,
|
| 1009 |
+
"loss": 0.3719,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.46265060240963857,
|
| 1014 |
+
"grad_norm": 2.1085941791534424,
|
| 1015 |
+
"learning_rate": 5.856817873664409e-06,
|
| 1016 |
+
"loss": 0.3306,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.46586345381526106,
|
| 1021 |
+
"grad_norm": 2.1606924533843994,
|
| 1022 |
+
"learning_rate": 5.8053581559707754e-06,
|
| 1023 |
+
"loss": 0.3847,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.46907630522088356,
|
| 1028 |
+
"grad_norm": 3.6136634349823,
|
| 1029 |
+
"learning_rate": 5.753810707520918e-06,
|
| 1030 |
+
"loss": 0.409,
|
| 1031 |
+
"step": 146
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.472289156626506,
|
| 1035 |
+
"grad_norm": 2.9589033126831055,
|
| 1036 |
+
"learning_rate": 5.702181143576323e-06,
|
| 1037 |
+
"loss": 0.519,
|
| 1038 |
+
"step": 147
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.4755020080321285,
|
| 1042 |
+
"grad_norm": 2.941072463989258,
|
| 1043 |
+
"learning_rate": 5.6504750883436275e-06,
|
| 1044 |
+
"loss": 0.3586,
|
| 1045 |
+
"step": 148
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.478714859437751,
|
| 1049 |
+
"grad_norm": 2.1762218475341797,
|
| 1050 |
+
"learning_rate": 5.5986981743619615e-06,
|
| 1051 |
+
"loss": 0.3692,
|
| 1052 |
+
"step": 149
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.4819277108433735,
|
| 1056 |
+
"grad_norm": 1.9385762214660645,
|
| 1057 |
+
"learning_rate": 5.546856041889374e-06,
|
| 1058 |
+
"loss": 0.3044,
|
| 1059 |
+
"step": 150
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.485140562248996,
|
| 1063 |
+
"grad_norm": 2.7357289791107178,
|
| 1064 |
+
"learning_rate": 5.494954338288404e-06,
|
| 1065 |
+
"loss": 0.4027,
|
| 1066 |
+
"step": 151
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.4883534136546185,
|
| 1070 |
+
"grad_norm": 2.4241161346435547,
|
| 1071 |
+
"learning_rate": 5.442998717410916e-06,
|
| 1072 |
+
"loss": 0.3591,
|
| 1073 |
+
"step": 152
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.491566265060241,
|
| 1077 |
+
"grad_norm": 5.211270809173584,
|
| 1078 |
+
"learning_rate": 5.390994838982178e-06,
|
| 1079 |
+
"loss": 0.3662,
|
| 1080 |
+
"step": 153
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.4947791164658635,
|
| 1084 |
+
"grad_norm": 2.5617518424987793,
|
| 1085 |
+
"learning_rate": 5.338948367984347e-06,
|
| 1086 |
+
"loss": 0.3067,
|
| 1087 |
+
"step": 154
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.4979919678714859,
|
| 1091 |
+
"grad_norm": 2.434497356414795,
|
| 1092 |
+
"learning_rate": 5.286864974039349e-06,
|
| 1093 |
+
"loss": 0.3023,
|
| 1094 |
+
"step": 155
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.5012048192771085,
|
| 1098 |
+
"grad_norm": 2.5631253719329834,
|
| 1099 |
+
"learning_rate": 5.234750330791268e-06,
|
| 1100 |
+
"loss": 0.3823,
|
| 1101 |
+
"step": 156
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.5044176706827309,
|
| 1105 |
+
"grad_norm": 2.003878593444824,
|
| 1106 |
+
"learning_rate": 5.182610115288296e-06,
|
| 1107 |
+
"loss": 0.361,
|
| 1108 |
+
"step": 157
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.5076305220883535,
|
| 1112 |
+
"grad_norm": 2.133862018585205,
|
| 1113 |
+
"learning_rate": 5.1304500073643045e-06,
|
| 1114 |
+
"loss": 0.2718,
|
| 1115 |
+
"step": 158
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.5108433734939759,
|
| 1119 |
+
"grad_norm": 2.344472646713257,
|
| 1120 |
+
"learning_rate": 5.078275689020129e-06,
|
| 1121 |
+
"loss": 0.2921,
|
| 1122 |
+
"step": 159
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.5140562248995983,
|
| 1126 |
+
"grad_norm": 2.4135284423828125,
|
| 1127 |
+
"learning_rate": 5.026092843804599e-06,
|
| 1128 |
+
"loss": 0.3341,
|
| 1129 |
+
"step": 160
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.5172690763052209,
|
| 1133 |
+
"grad_norm": 2.5881218910217285,
|
| 1134 |
+
"learning_rate": 4.973907156195405e-06,
|
| 1135 |
+
"loss": 0.2986,
|
| 1136 |
+
"step": 161
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.5204819277108433,
|
| 1140 |
+
"grad_norm": 2.7781827449798584,
|
| 1141 |
+
"learning_rate": 4.921724310979872e-06,
|
| 1142 |
+
"loss": 0.4795,
|
| 1143 |
+
"step": 162
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.5236947791164659,
|
| 1147 |
+
"grad_norm": 2.816742181777954,
|
| 1148 |
+
"learning_rate": 4.869549992635697e-06,
|
| 1149 |
+
"loss": 0.4024,
|
| 1150 |
+
"step": 163
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.5269076305220883,
|
| 1154 |
+
"grad_norm": 2.479278326034546,
|
| 1155 |
+
"learning_rate": 4.817389884711706e-06,
|
| 1156 |
+
"loss": 0.381,
|
| 1157 |
+
"step": 164
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.5301204819277109,
|
| 1161 |
+
"grad_norm": 2.0837013721466064,
|
| 1162 |
+
"learning_rate": 4.765249669208733e-06,
|
| 1163 |
+
"loss": 0.2775,
|
| 1164 |
+
"step": 165
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.5333333333333333,
|
| 1168 |
+
"grad_norm": 3.396789312362671,
|
| 1169 |
+
"learning_rate": 4.713135025960652e-06,
|
| 1170 |
+
"loss": 0.3469,
|
| 1171 |
+
"step": 166
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.5365461847389559,
|
| 1175 |
+
"grad_norm": 2.179645299911499,
|
| 1176 |
+
"learning_rate": 4.661051632015655e-06,
|
| 1177 |
+
"loss": 0.375,
|
| 1178 |
+
"step": 167
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.5397590361445783,
|
| 1182 |
+
"grad_norm": 2.4666926860809326,
|
| 1183 |
+
"learning_rate": 4.609005161017824e-06,
|
| 1184 |
+
"loss": 0.3723,
|
| 1185 |
+
"step": 168
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.5429718875502008,
|
| 1189 |
+
"grad_norm": 2.0646708011627197,
|
| 1190 |
+
"learning_rate": 4.557001282589086e-06,
|
| 1191 |
+
"loss": 0.3036,
|
| 1192 |
+
"step": 169
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.5461847389558233,
|
| 1196 |
+
"grad_norm": 2.0979270935058594,
|
| 1197 |
+
"learning_rate": 4.505045661711596e-06,
|
| 1198 |
+
"loss": 0.312,
|
| 1199 |
+
"step": 170
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.5493975903614458,
|
| 1203 |
+
"grad_norm": 2.0291059017181396,
|
| 1204 |
+
"learning_rate": 4.4531439581106295e-06,
|
| 1205 |
+
"loss": 0.3151,
|
| 1206 |
+
"step": 171
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.5526104417670683,
|
| 1210 |
+
"grad_norm": 1.8637864589691162,
|
| 1211 |
+
"learning_rate": 4.401301825638039e-06,
|
| 1212 |
+
"loss": 0.2561,
|
| 1213 |
+
"step": 172
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.5558232931726907,
|
| 1217 |
+
"grad_norm": 2.0765888690948486,
|
| 1218 |
+
"learning_rate": 4.349524911656373e-06,
|
| 1219 |
+
"loss": 0.3333,
|
| 1220 |
+
"step": 173
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.5590361445783133,
|
| 1224 |
+
"grad_norm": 2.5530149936676025,
|
| 1225 |
+
"learning_rate": 4.297818856423679e-06,
|
| 1226 |
+
"loss": 0.3819,
|
| 1227 |
+
"step": 174
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.5622489959839357,
|
| 1231 |
+
"grad_norm": 2.2014477252960205,
|
| 1232 |
+
"learning_rate": 4.2461892924790825e-06,
|
| 1233 |
+
"loss": 0.254,
|
| 1234 |
+
"step": 175
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.5654618473895582,
|
| 1238 |
+
"grad_norm": 3.6043734550476074,
|
| 1239 |
+
"learning_rate": 4.194641844029227e-06,
|
| 1240 |
+
"loss": 0.4335,
|
| 1241 |
+
"step": 176
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.5686746987951807,
|
| 1245 |
+
"grad_norm": 2.4686837196350098,
|
| 1246 |
+
"learning_rate": 4.143182126335594e-06,
|
| 1247 |
+
"loss": 0.2313,
|
| 1248 |
+
"step": 177
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.5718875502008032,
|
| 1252 |
+
"grad_norm": 1.9996862411499023,
|
| 1253 |
+
"learning_rate": 4.091815745102818e-06,
|
| 1254 |
+
"loss": 0.296,
|
| 1255 |
+
"step": 178
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.5751004016064257,
|
| 1259 |
+
"grad_norm": 2.37868595123291,
|
| 1260 |
+
"learning_rate": 4.040548295868039e-06,
|
| 1261 |
+
"loss": 0.2864,
|
| 1262 |
+
"step": 179
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.5783132530120482,
|
| 1266 |
+
"grad_norm": 2.2569093704223633,
|
| 1267 |
+
"learning_rate": 3.9893853633913485e-06,
|
| 1268 |
+
"loss": 0.2444,
|
| 1269 |
+
"step": 180
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.5815261044176707,
|
| 1273 |
+
"grad_norm": 3.60836124420166,
|
| 1274 |
+
"learning_rate": 3.938332521047434e-06,
|
| 1275 |
+
"loss": 0.3296,
|
| 1276 |
+
"step": 181
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.5847389558232932,
|
| 1280 |
+
"grad_norm": 2.706819772720337,
|
| 1281 |
+
"learning_rate": 3.887395330218429e-06,
|
| 1282 |
+
"loss": 0.1887,
|
| 1283 |
+
"step": 182
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.5879518072289157,
|
| 1287 |
+
"grad_norm": 2.409383535385132,
|
| 1288 |
+
"learning_rate": 3.836579339688116e-06,
|
| 1289 |
+
"loss": 0.29,
|
| 1290 |
+
"step": 183
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.5911646586345382,
|
| 1294 |
+
"grad_norm": 2.9502406120300293,
|
| 1295 |
+
"learning_rate": 3.7858900850374596e-06,
|
| 1296 |
+
"loss": 0.3951,
|
| 1297 |
+
"step": 184
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.5943775100401606,
|
| 1301 |
+
"grad_norm": 2.052128314971924,
|
| 1302 |
+
"learning_rate": 3.7353330880415963e-06,
|
| 1303 |
+
"loss": 0.3078,
|
| 1304 |
+
"step": 185
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.5975903614457831,
|
| 1308 |
+
"grad_norm": 2.404327630996704,
|
| 1309 |
+
"learning_rate": 3.6849138560683305e-06,
|
| 1310 |
+
"loss": 0.4112,
|
| 1311 |
+
"step": 186
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.6008032128514056,
|
| 1315 |
+
"grad_norm": 2.3815231323242188,
|
| 1316 |
+
"learning_rate": 3.634637881478196e-06,
|
| 1317 |
+
"loss": 0.322,
|
| 1318 |
+
"step": 187
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.6040160642570281,
|
| 1322 |
+
"grad_norm": 2.2108402252197266,
|
| 1323 |
+
"learning_rate": 3.5845106410261417e-06,
|
| 1324 |
+
"loss": 0.3744,
|
| 1325 |
+
"step": 188
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.6072289156626506,
|
| 1329 |
+
"grad_norm": 2.1574432849884033,
|
| 1330 |
+
"learning_rate": 3.534537595264944e-06,
|
| 1331 |
+
"loss": 0.2951,
|
| 1332 |
+
"step": 189
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.6104417670682731,
|
| 1336 |
+
"grad_norm": 2.283750057220459,
|
| 1337 |
+
"learning_rate": 3.4847241879503574e-06,
|
| 1338 |
+
"loss": 0.2728,
|
| 1339 |
+
"step": 190
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.6136546184738956,
|
| 1343 |
+
"grad_norm": 2.0512266159057617,
|
| 1344 |
+
"learning_rate": 3.435075845448105e-06,
|
| 1345 |
+
"loss": 0.2585,
|
| 1346 |
+
"step": 191
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.6168674698795181,
|
| 1350 |
+
"grad_norm": 2.368990182876587,
|
| 1351 |
+
"learning_rate": 3.3855979761427705e-06,
|
| 1352 |
+
"loss": 0.4226,
|
| 1353 |
+
"step": 192
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.6200803212851406,
|
| 1357 |
+
"grad_norm": 2.43058705329895,
|
| 1358 |
+
"learning_rate": 3.3362959698486307e-06,
|
| 1359 |
+
"loss": 0.3779,
|
| 1360 |
+
"step": 193
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.623293172690763,
|
| 1364 |
+
"grad_norm": 2.8462181091308594,
|
| 1365 |
+
"learning_rate": 3.287175197222537e-06,
|
| 1366 |
+
"loss": 0.4504,
|
| 1367 |
+
"step": 194
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.6265060240963856,
|
| 1371 |
+
"grad_norm": 2.358142614364624,
|
| 1372 |
+
"learning_rate": 3.2382410091788567e-06,
|
| 1373 |
+
"loss": 0.3824,
|
| 1374 |
+
"step": 195
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.629718875502008,
|
| 1378 |
+
"grad_norm": 1.9749348163604736,
|
| 1379 |
+
"learning_rate": 3.189498736306584e-06,
|
| 1380 |
+
"loss": 0.3309,
|
| 1381 |
+
"step": 196
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.6329317269076306,
|
| 1385 |
+
"grad_norm": 2.166022300720215,
|
| 1386 |
+
"learning_rate": 3.140953688288658e-06,
|
| 1387 |
+
"loss": 0.1816,
|
| 1388 |
+
"step": 197
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.636144578313253,
|
| 1392 |
+
"grad_norm": 2.078129291534424,
|
| 1393 |
+
"learning_rate": 3.0926111533235526e-06,
|
| 1394 |
+
"loss": 0.3417,
|
| 1395 |
+
"step": 198
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.6393574297188755,
|
| 1399 |
+
"grad_norm": 2.0340514183044434,
|
| 1400 |
+
"learning_rate": 3.044476397549221e-06,
|
| 1401 |
+
"loss": 0.2881,
|
| 1402 |
+
"step": 199
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.642570281124498,
|
| 1406 |
+
"grad_norm": 2.3322885036468506,
|
| 1407 |
+
"learning_rate": 2.9965546644694287e-06,
|
| 1408 |
+
"loss": 0.2418,
|
| 1409 |
+
"step": 200
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.6457831325301204,
|
| 1413 |
+
"grad_norm": 2.1144306659698486,
|
| 1414 |
+
"learning_rate": 2.948851174382565e-06,
|
| 1415 |
+
"loss": 0.2758,
|
| 1416 |
+
"step": 201
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.648995983935743,
|
| 1420 |
+
"grad_norm": 2.089343309402466,
|
| 1421 |
+
"learning_rate": 2.9013711238129693e-06,
|
| 1422 |
+
"loss": 0.3059,
|
| 1423 |
+
"step": 202
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.6522088353413654,
|
| 1427 |
+
"grad_norm": 2.4294567108154297,
|
| 1428 |
+
"learning_rate": 2.8541196849448582e-06,
|
| 1429 |
+
"loss": 0.3653,
|
| 1430 |
+
"step": 203
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.655421686746988,
|
| 1434 |
+
"grad_norm": 2.432567596435547,
|
| 1435 |
+
"learning_rate": 2.8071020050588927e-06,
|
| 1436 |
+
"loss": 0.2259,
|
| 1437 |
+
"step": 204
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.6586345381526104,
|
| 1441 |
+
"grad_norm": 2.2744345664978027,
|
| 1442 |
+
"learning_rate": 2.760323205971476e-06,
|
| 1443 |
+
"loss": 0.3401,
|
| 1444 |
+
"step": 205
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.661847389558233,
|
| 1448 |
+
"grad_norm": 2.8238182067871094,
|
| 1449 |
+
"learning_rate": 2.7137883834768076e-06,
|
| 1450 |
+
"loss": 0.2912,
|
| 1451 |
+
"step": 206
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.6650602409638554,
|
| 1455 |
+
"grad_norm": 2.3627312183380127,
|
| 1456 |
+
"learning_rate": 2.6675026067917808e-06,
|
| 1457 |
+
"loss": 0.2491,
|
| 1458 |
+
"step": 207
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.668273092369478,
|
| 1462 |
+
"grad_norm": 2.5185694694519043,
|
| 1463 |
+
"learning_rate": 2.621470918003768e-06,
|
| 1464 |
+
"loss": 0.2879,
|
| 1465 |
+
"step": 208
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.6714859437751004,
|
| 1469 |
+
"grad_norm": 2.401811361312866,
|
| 1470 |
+
"learning_rate": 2.5756983315213748e-06,
|
| 1471 |
+
"loss": 0.3885,
|
| 1472 |
+
"step": 209
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.6746987951807228,
|
| 1476 |
+
"grad_norm": 2.2231993675231934,
|
| 1477 |
+
"learning_rate": 2.5301898335281994e-06,
|
| 1478 |
+
"loss": 0.2833,
|
| 1479 |
+
"step": 210
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.6779116465863454,
|
| 1483 |
+
"grad_norm": 2.2760813236236572,
|
| 1484 |
+
"learning_rate": 2.4849503814396624e-06,
|
| 1485 |
+
"loss": 0.2532,
|
| 1486 |
+
"step": 211
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.6811244979919678,
|
| 1490 |
+
"grad_norm": 2.149466037750244,
|
| 1491 |
+
"learning_rate": 2.439984903362988e-06,
|
| 1492 |
+
"loss": 0.2805,
|
| 1493 |
+
"step": 212
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.6843373493975904,
|
| 1497 |
+
"grad_norm": 2.3873984813690186,
|
| 1498 |
+
"learning_rate": 2.3952982975603494e-06,
|
| 1499 |
+
"loss": 0.3158,
|
| 1500 |
+
"step": 213
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.6875502008032128,
|
| 1504 |
+
"grad_norm": 2.5702414512634277,
|
| 1505 |
+
"learning_rate": 2.3508954319153e-06,
|
| 1506 |
+
"loss": 0.3221,
|
| 1507 |
+
"step": 214
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.6907630522088354,
|
| 1511 |
+
"grad_norm": 2.643507719039917,
|
| 1512 |
+
"learning_rate": 2.306781143402485e-06,
|
| 1513 |
+
"loss": 0.3544,
|
| 1514 |
+
"step": 215
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.6939759036144578,
|
| 1518 |
+
"grad_norm": 2.201829671859741,
|
| 1519 |
+
"learning_rate": 2.2629602375607373e-06,
|
| 1520 |
+
"loss": 0.3529,
|
| 1521 |
+
"step": 216
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.6971887550200804,
|
| 1525 |
+
"grad_norm": 2.1155354976654053,
|
| 1526 |
+
"learning_rate": 2.219437487969588e-06,
|
| 1527 |
+
"loss": 0.2572,
|
| 1528 |
+
"step": 217
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.7004016064257028,
|
| 1532 |
+
"grad_norm": 2.038470506668091,
|
| 1533 |
+
"learning_rate": 2.1762176357292582e-06,
|
| 1534 |
+
"loss": 0.2295,
|
| 1535 |
+
"step": 218
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.7036144578313253,
|
| 1539 |
+
"grad_norm": 2.023308753967285,
|
| 1540 |
+
"learning_rate": 2.1333053889442033e-06,
|
| 1541 |
+
"loss": 0.3592,
|
| 1542 |
+
"step": 219
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.7068273092369478,
|
| 1546 |
+
"grad_norm": 2.5169191360473633,
|
| 1547 |
+
"learning_rate": 2.0907054222102367e-06,
|
| 1548 |
+
"loss": 0.3218,
|
| 1549 |
+
"step": 220
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.7100401606425703,
|
| 1553 |
+
"grad_norm": 2.5444135665893555,
|
| 1554 |
+
"learning_rate": 2.048422376105299e-06,
|
| 1555 |
+
"loss": 0.3331,
|
| 1556 |
+
"step": 221
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.7132530120481928,
|
| 1560 |
+
"grad_norm": 2.6340672969818115,
|
| 1561 |
+
"learning_rate": 2.0064608566839584e-06,
|
| 1562 |
+
"loss": 0.4002,
|
| 1563 |
+
"step": 222
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.7164658634538152,
|
| 1567 |
+
"grad_norm": 2.0738048553466797,
|
| 1568 |
+
"learning_rate": 1.964825434975639e-06,
|
| 1569 |
+
"loss": 0.2908,
|
| 1570 |
+
"step": 223
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.7196787148594378,
|
| 1574 |
+
"grad_norm": 2.0317463874816895,
|
| 1575 |
+
"learning_rate": 1.923520646486695e-06,
|
| 1576 |
+
"loss": 0.2293,
|
| 1577 |
+
"step": 224
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.7228915662650602,
|
| 1581 |
+
"grad_norm": 2.2780821323394775,
|
| 1582 |
+
"learning_rate": 1.8825509907063328e-06,
|
| 1583 |
+
"loss": 0.2623,
|
| 1584 |
+
"step": 225
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.7261044176706827,
|
| 1588 |
+
"grad_norm": 2.3251006603240967,
|
| 1589 |
+
"learning_rate": 1.8419209306164653e-06,
|
| 1590 |
+
"loss": 0.2582,
|
| 1591 |
+
"step": 226
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.7293172690763052,
|
| 1595 |
+
"grad_norm": 2.357585906982422,
|
| 1596 |
+
"learning_rate": 1.8016348922055448e-06,
|
| 1597 |
+
"loss": 0.2822,
|
| 1598 |
+
"step": 227
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.7325301204819277,
|
| 1602 |
+
"grad_norm": 2.6229782104492188,
|
| 1603 |
+
"learning_rate": 1.7616972639864166e-06,
|
| 1604 |
+
"loss": 0.3912,
|
| 1605 |
+
"step": 228
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.7357429718875502,
|
| 1609 |
+
"grad_norm": 2.2909278869628906,
|
| 1610 |
+
"learning_rate": 1.7221123965182712e-06,
|
| 1611 |
+
"loss": 0.3465,
|
| 1612 |
+
"step": 229
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.7389558232931727,
|
| 1616 |
+
"grad_norm": 1.9987268447875977,
|
| 1617 |
+
"learning_rate": 1.6828846019327128e-06,
|
| 1618 |
+
"loss": 0.3453,
|
| 1619 |
+
"step": 230
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.7421686746987952,
|
| 1623 |
+
"grad_norm": 2.4425132274627686,
|
| 1624 |
+
"learning_rate": 1.6440181534640277e-06,
|
| 1625 |
+
"loss": 0.2378,
|
| 1626 |
+
"step": 231
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.7453815261044177,
|
| 1630 |
+
"grad_norm": 1.847144365310669,
|
| 1631 |
+
"learning_rate": 1.6055172849836826e-06,
|
| 1632 |
+
"loss": 0.2161,
|
| 1633 |
+
"step": 232
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.7485943775100402,
|
| 1637 |
+
"grad_norm": 2.3810346126556396,
|
| 1638 |
+
"learning_rate": 1.567386190539107e-06,
|
| 1639 |
+
"loss": 0.2798,
|
| 1640 |
+
"step": 233
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.7518072289156627,
|
| 1644 |
+
"grad_norm": 1.8688887357711792,
|
| 1645 |
+
"learning_rate": 1.5296290238968303e-06,
|
| 1646 |
+
"loss": 0.2168,
|
| 1647 |
+
"step": 234
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.7550200803212851,
|
| 1651 |
+
"grad_norm": 3.1742520332336426,
|
| 1652 |
+
"learning_rate": 1.4922498980899907e-06,
|
| 1653 |
+
"loss": 0.3661,
|
| 1654 |
+
"step": 235
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.7582329317269076,
|
| 1658 |
+
"grad_norm": 2.355210542678833,
|
| 1659 |
+
"learning_rate": 1.4552528849702852e-06,
|
| 1660 |
+
"loss": 0.2874,
|
| 1661 |
+
"step": 236
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.7614457831325301,
|
| 1665 |
+
"grad_norm": 3.493511915206909,
|
| 1666 |
+
"learning_rate": 1.4186420147644053e-06,
|
| 1667 |
+
"loss": 0.3802,
|
| 1668 |
+
"step": 237
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.7646586345381526,
|
| 1672 |
+
"grad_norm": 1.8001739978790283,
|
| 1673 |
+
"learning_rate": 1.3824212756350196e-06,
|
| 1674 |
+
"loss": 0.2334,
|
| 1675 |
+
"step": 238
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.7678714859437751,
|
| 1679 |
+
"grad_norm": 1.9317985773086548,
|
| 1680 |
+
"learning_rate": 1.3465946132463125e-06,
|
| 1681 |
+
"loss": 0.245,
|
| 1682 |
+
"step": 239
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.7710843373493976,
|
| 1686 |
+
"grad_norm": 2.0644335746765137,
|
| 1687 |
+
"learning_rate": 1.3111659303341824e-06,
|
| 1688 |
+
"loss": 0.243,
|
| 1689 |
+
"step": 240
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.7742971887550201,
|
| 1693 |
+
"grad_norm": 2.091796398162842,
|
| 1694 |
+
"learning_rate": 1.2761390862810907e-06,
|
| 1695 |
+
"loss": 0.2382,
|
| 1696 |
+
"step": 241
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.7775100401606426,
|
| 1700 |
+
"grad_norm": 2.3172249794006348,
|
| 1701 |
+
"learning_rate": 1.2415178966956531e-06,
|
| 1702 |
+
"loss": 0.2797,
|
| 1703 |
+
"step": 242
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.7807228915662651,
|
| 1707 |
+
"grad_norm": 2.269488573074341,
|
| 1708 |
+
"learning_rate": 1.2073061329969843e-06,
|
| 1709 |
+
"loss": 0.3524,
|
| 1710 |
+
"step": 243
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.7839357429718875,
|
| 1714 |
+
"grad_norm": 2.3127260208129883,
|
| 1715 |
+
"learning_rate": 1.1735075220038634e-06,
|
| 1716 |
+
"loss": 0.2378,
|
| 1717 |
+
"step": 244
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.7871485943775101,
|
| 1721 |
+
"grad_norm": 1.958678126335144,
|
| 1722 |
+
"learning_rate": 1.1401257455287612e-06,
|
| 1723 |
+
"loss": 0.2539,
|
| 1724 |
+
"step": 245
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.7903614457831325,
|
| 1728 |
+
"grad_norm": 2.530526876449585,
|
| 1729 |
+
"learning_rate": 1.107164439976764e-06,
|
| 1730 |
+
"loss": 0.2905,
|
| 1731 |
+
"step": 246
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.793574297188755,
|
| 1735 |
+
"grad_norm": 2.522650957107544,
|
| 1736 |
+
"learning_rate": 1.0746271959494453e-06,
|
| 1737 |
+
"loss": 0.3653,
|
| 1738 |
+
"step": 247
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.7967871485943775,
|
| 1742 |
+
"grad_norm": 2.9361941814422607,
|
| 1743 |
+
"learning_rate": 1.04251755785373e-06,
|
| 1744 |
+
"loss": 0.3054,
|
| 1745 |
+
"step": 248
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.8,
|
| 1749 |
+
"grad_norm": 2.346240758895874,
|
| 1750 |
+
"learning_rate": 1.0108390235157828e-06,
|
| 1751 |
+
"loss": 0.2623,
|
| 1752 |
+
"step": 249
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.8032128514056225,
|
| 1756 |
+
"grad_norm": 2.452631711959839,
|
| 1757 |
+
"learning_rate": 9.795950437999852e-07,
|
| 1758 |
+
"loss": 0.2113,
|
| 1759 |
+
"step": 250
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.8064257028112449,
|
| 1763 |
+
"grad_norm": 2.9312245845794678,
|
| 1764 |
+
"learning_rate": 9.487890222330137e-07,
|
| 1765 |
+
"loss": 0.4234,
|
| 1766 |
+
"step": 251
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.8096385542168675,
|
| 1770 |
+
"grad_norm": 2.1248555183410645,
|
| 1771 |
+
"learning_rate": 9.184243146330829e-07,
|
| 1772 |
+
"loss": 0.2872,
|
| 1773 |
+
"step": 252
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.8128514056224899,
|
| 1777 |
+
"grad_norm": 2.393862247467041,
|
| 1778 |
+
"learning_rate": 8.885042287443785e-07,
|
| 1779 |
+
"loss": 0.2849,
|
| 1780 |
+
"step": 253
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.8160642570281125,
|
| 1784 |
+
"grad_norm": 2.453737258911133,
|
| 1785 |
+
"learning_rate": 8.590320238767425e-07,
|
| 1786 |
+
"loss": 0.2962,
|
| 1787 |
+
"step": 254
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.8192771084337349,
|
| 1791 |
+
"grad_norm": 2.9770538806915283,
|
| 1792 |
+
"learning_rate": 8.30010910550611e-07,
|
| 1793 |
+
"loss": 0.3318,
|
| 1794 |
+
"step": 255
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.8224899598393575,
|
| 1798 |
+
"grad_norm": 1.8912490606307983,
|
| 1799 |
+
"learning_rate": 8.014440501472909e-07,
|
| 1800 |
+
"loss": 0.2446,
|
| 1801 |
+
"step": 256
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.8257028112449799,
|
| 1805 |
+
"grad_norm": 3.3345978260040283,
|
| 1806 |
+
"learning_rate": 7.733345545645726e-07,
|
| 1807 |
+
"loss": 0.3895,
|
| 1808 |
+
"step": 257
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.8289156626506025,
|
| 1812 |
+
"grad_norm": 2.2788500785827637,
|
| 1813 |
+
"learning_rate": 7.456854858777418e-07,
|
| 1814 |
+
"loss": 0.2767,
|
| 1815 |
+
"step": 258
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.8321285140562249,
|
| 1819 |
+
"grad_norm": 2.039923906326294,
|
| 1820 |
+
"learning_rate": 7.184998560060114e-07,
|
| 1821 |
+
"loss": 0.1987,
|
| 1822 |
+
"step": 259
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.8353413654618473,
|
| 1826 |
+
"grad_norm": 2.3319668769836426,
|
| 1827 |
+
"learning_rate": 6.917806263844268e-07,
|
| 1828 |
+
"loss": 0.2265,
|
| 1829 |
+
"step": 260
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.8385542168674699,
|
| 1833 |
+
"grad_norm": 3.085174560546875,
|
| 1834 |
+
"learning_rate": 6.655307076412637e-07,
|
| 1835 |
+
"loss": 0.3313,
|
| 1836 |
+
"step": 261
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.8417670682730923,
|
| 1840 |
+
"grad_norm": 1.9811396598815918,
|
| 1841 |
+
"learning_rate": 6.397529592809615e-07,
|
| 1842 |
+
"loss": 0.2447,
|
| 1843 |
+
"step": 262
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.8449799196787149,
|
| 1847 |
+
"grad_norm": 1.8202673196792603,
|
| 1848 |
+
"learning_rate": 6.14450189372628e-07,
|
| 1849 |
+
"loss": 0.2427,
|
| 1850 |
+
"step": 263
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.8481927710843373,
|
| 1854 |
+
"grad_norm": 2.299744129180908,
|
| 1855 |
+
"learning_rate": 5.896251542441395e-07,
|
| 1856 |
+
"loss": 0.2412,
|
| 1857 |
+
"step": 264
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.8514056224899599,
|
| 1861 |
+
"grad_norm": 2.096524953842163,
|
| 1862 |
+
"learning_rate": 5.652805581818943e-07,
|
| 1863 |
+
"loss": 0.2329,
|
| 1864 |
+
"step": 265
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.8546184738955823,
|
| 1868 |
+
"grad_norm": 2.4148600101470947,
|
| 1869 |
+
"learning_rate": 5.414190531362162e-07,
|
| 1870 |
+
"loss": 0.2183,
|
| 1871 |
+
"step": 266
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.8578313253012049,
|
| 1875 |
+
"grad_norm": 1.961333155632019,
|
| 1876 |
+
"learning_rate": 5.180432384324691e-07,
|
| 1877 |
+
"loss": 0.2278,
|
| 1878 |
+
"step": 267
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.8610441767068273,
|
| 1882 |
+
"grad_norm": 1.9496490955352783,
|
| 1883 |
+
"learning_rate": 4.951556604879049e-07,
|
| 1884 |
+
"loss": 0.1662,
|
| 1885 |
+
"step": 268
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.8642570281124498,
|
| 1889 |
+
"grad_norm": 2.195952892303467,
|
| 1890 |
+
"learning_rate": 4.727588125342669e-07,
|
| 1891 |
+
"loss": 0.2645,
|
| 1892 |
+
"step": 269
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.8674698795180723,
|
| 1896 |
+
"grad_norm": 1.8171138763427734,
|
| 1897 |
+
"learning_rate": 4.508551343462014e-07,
|
| 1898 |
+
"loss": 0.1446,
|
| 1899 |
+
"step": 270
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.8706827309236947,
|
| 1903 |
+
"grad_norm": 2.3251218795776367,
|
| 1904 |
+
"learning_rate": 4.29447011975474e-07,
|
| 1905 |
+
"loss": 0.2679,
|
| 1906 |
+
"step": 271
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.8738955823293173,
|
| 1910 |
+
"grad_norm": 2.06396746635437,
|
| 1911 |
+
"learning_rate": 4.0853677749105426e-07,
|
| 1912 |
+
"loss": 0.2793,
|
| 1913 |
+
"step": 272
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.8771084337349397,
|
| 1917 |
+
"grad_norm": 2.0605921745300293,
|
| 1918 |
+
"learning_rate": 3.8812670872507454e-07,
|
| 1919 |
+
"loss": 0.2798,
|
| 1920 |
+
"step": 273
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.8803212851405623,
|
| 1924 |
+
"grad_norm": 2.0419671535491943,
|
| 1925 |
+
"learning_rate": 3.6821902902469066e-07,
|
| 1926 |
+
"loss": 0.3494,
|
| 1927 |
+
"step": 274
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.8835341365461847,
|
| 1931 |
+
"grad_norm": 1.9494353532791138,
|
| 1932 |
+
"learning_rate": 3.4881590700989175e-07,
|
| 1933 |
+
"loss": 0.3164,
|
| 1934 |
+
"step": 275
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.8867469879518072,
|
| 1938 |
+
"grad_norm": 2.2515294551849365,
|
| 1939 |
+
"learning_rate": 3.299194563372604e-07,
|
| 1940 |
+
"loss": 0.2364,
|
| 1941 |
+
"step": 276
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.8899598393574297,
|
| 1945 |
+
"grad_norm": 2.4932332038879395,
|
| 1946 |
+
"learning_rate": 3.1153173546972395e-07,
|
| 1947 |
+
"loss": 0.1982,
|
| 1948 |
+
"step": 277
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.8931726907630522,
|
| 1952 |
+
"grad_norm": 1.9155004024505615,
|
| 1953 |
+
"learning_rate": 2.9365474745231935e-07,
|
| 1954 |
+
"loss": 0.2864,
|
| 1955 |
+
"step": 278
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.8963855421686747,
|
| 1959 |
+
"grad_norm": 2.346914529800415,
|
| 1960 |
+
"learning_rate": 2.7629043969399193e-07,
|
| 1961 |
+
"loss": 0.3323,
|
| 1962 |
+
"step": 279
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.8995983935742972,
|
| 1966 |
+
"grad_norm": 1.957244634628296,
|
| 1967 |
+
"learning_rate": 2.594407037554586e-07,
|
| 1968 |
+
"loss": 0.1917,
|
| 1969 |
+
"step": 280
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.9028112449799197,
|
| 1973 |
+
"grad_norm": 2.4275577068328857,
|
| 1974 |
+
"learning_rate": 2.431073751431529e-07,
|
| 1975 |
+
"loss": 0.3431,
|
| 1976 |
+
"step": 281
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.9060240963855422,
|
| 1980 |
+
"grad_norm": 2.3854384422302246,
|
| 1981 |
+
"learning_rate": 2.2729223310927473e-07,
|
| 1982 |
+
"loss": 0.1988,
|
| 1983 |
+
"step": 282
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.9092369477911647,
|
| 1987 |
+
"grad_norm": 2.0022900104522705,
|
| 1988 |
+
"learning_rate": 2.1199700045797077e-07,
|
| 1989 |
+
"loss": 0.2806,
|
| 1990 |
+
"step": 283
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.9124497991967871,
|
| 1994 |
+
"grad_norm": 2.0023062229156494,
|
| 1995 |
+
"learning_rate": 1.9722334335766092e-07,
|
| 1996 |
+
"loss": 0.3024,
|
| 1997 |
+
"step": 284
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.9156626506024096,
|
| 2001 |
+
"grad_norm": 1.8861989974975586,
|
| 2002 |
+
"learning_rate": 1.829728711595391e-07,
|
| 2003 |
+
"loss": 0.2559,
|
| 2004 |
+
"step": 285
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.9188755020080321,
|
| 2008 |
+
"grad_norm": 2.1966798305511475,
|
| 2009 |
+
"learning_rate": 1.6924713622225975e-07,
|
| 2010 |
+
"loss": 0.2724,
|
| 2011 |
+
"step": 286
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.9220883534136546,
|
| 2015 |
+
"grad_norm": 1.6553095579147339,
|
| 2016 |
+
"learning_rate": 1.5604763374283073e-07,
|
| 2017 |
+
"loss": 0.1017,
|
| 2018 |
+
"step": 287
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.9253012048192771,
|
| 2022 |
+
"grad_norm": 2.2511794567108154,
|
| 2023 |
+
"learning_rate": 1.4337580159373864e-07,
|
| 2024 |
+
"loss": 0.2722,
|
| 2025 |
+
"step": 288
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.9285140562248996,
|
| 2029 |
+
"grad_norm": 2.169060230255127,
|
| 2030 |
+
"learning_rate": 1.3123302016631477e-07,
|
| 2031 |
+
"loss": 0.265,
|
| 2032 |
+
"step": 289
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.9317269076305221,
|
| 2036 |
+
"grad_norm": 2.9961092472076416,
|
| 2037 |
+
"learning_rate": 1.196206122203647e-07,
|
| 2038 |
+
"loss": 0.3931,
|
| 2039 |
+
"step": 290
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.9349397590361446,
|
| 2043 |
+
"grad_norm": 2.435908794403076,
|
| 2044 |
+
"learning_rate": 1.0853984274007246e-07,
|
| 2045 |
+
"loss": 0.2906,
|
| 2046 |
+
"step": 291
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.9381526104417671,
|
| 2050 |
+
"grad_norm": 2.8427674770355225,
|
| 2051 |
+
"learning_rate": 9.799191879620474e-08,
|
| 2052 |
+
"loss": 0.2227,
|
| 2053 |
+
"step": 292
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.9413654618473896,
|
| 2057 |
+
"grad_norm": 2.3079280853271484,
|
| 2058 |
+
"learning_rate": 8.797798941461655e-08,
|
| 2059 |
+
"loss": 0.2599,
|
| 2060 |
+
"step": 293
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.944578313253012,
|
| 2064 |
+
"grad_norm": 2.447566032409668,
|
| 2065 |
+
"learning_rate": 7.84991454510864e-08,
|
| 2066 |
+
"loss": 0.25,
|
| 2067 |
+
"step": 294
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.9477911646586346,
|
| 2071 |
+
"grad_norm": 2.5724740028381348,
|
| 2072 |
+
"learning_rate": 6.955641947248127e-08,
|
| 2073 |
+
"loss": 0.236,
|
| 2074 |
+
"step": 295
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.951004016064257,
|
| 2078 |
+
"grad_norm": 2.4259979724884033,
|
| 2079 |
+
"learning_rate": 6.115078564427946e-08,
|
| 2080 |
+
"loss": 0.2004,
|
| 2081 |
+
"step": 296
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.9542168674698795,
|
| 2085 |
+
"grad_norm": 1.9986400604248047,
|
| 2086 |
+
"learning_rate": 5.3283159624448745e-08,
|
| 2087 |
+
"loss": 0.2437,
|
| 2088 |
+
"step": 297
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.957429718875502,
|
| 2092 |
+
"grad_norm": 2.7118661403656006,
|
| 2093 |
+
"learning_rate": 4.5954398463700647e-08,
|
| 2094 |
+
"loss": 0.4067,
|
| 2095 |
+
"step": 298
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.9606425702811245,
|
| 2099 |
+
"grad_norm": 2.030395984649658,
|
| 2100 |
+
"learning_rate": 3.916530051212841e-08,
|
| 2101 |
+
"loss": 0.2347,
|
| 2102 |
+
"step": 299
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.963855421686747,
|
| 2106 |
+
"grad_norm": 1.5845026969909668,
|
| 2107 |
+
"learning_rate": 3.2916605332238284e-08,
|
| 2108 |
+
"loss": 0.153,
|
| 2109 |
+
"step": 300
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.9670682730923694,
|
| 2113 |
+
"grad_norm": 2.0269935131073,
|
| 2114 |
+
"learning_rate": 2.7208993618390578e-08,
|
| 2115 |
+
"loss": 0.1627,
|
| 2116 |
+
"step": 301
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.970281124497992,
|
| 2120 |
+
"grad_norm": 1.7263789176940918,
|
| 2121 |
+
"learning_rate": 2.2043087122644023e-08,
|
| 2122 |
+
"loss": 0.1601,
|
| 2123 |
+
"step": 302
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.9734939759036144,
|
| 2127 |
+
"grad_norm": 1.7540363073349,
|
| 2128 |
+
"learning_rate": 1.741944858702771e-08,
|
| 2129 |
+
"loss": 0.2004,
|
| 2130 |
+
"step": 303
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.976706827309237,
|
| 2134 |
+
"grad_norm": 2.1071643829345703,
|
| 2135 |
+
"learning_rate": 1.333858168224178e-08,
|
| 2136 |
+
"loss": 0.2089,
|
| 2137 |
+
"step": 304
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.9799196787148594,
|
| 2141 |
+
"grad_norm": 2.682478904724121,
|
| 2142 |
+
"learning_rate": 9.800930952786336e-09,
|
| 2143 |
+
"loss": 0.3269,
|
| 2144 |
+
"step": 305
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.983132530120482,
|
| 2148 |
+
"grad_norm": 2.5605387687683105,
|
| 2149 |
+
"learning_rate": 6.806881768539053e-09,
|
| 2150 |
+
"loss": 0.2146,
|
| 2151 |
+
"step": 306
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.9863453815261044,
|
| 2155 |
+
"grad_norm": 1.8436514139175415,
|
| 2156 |
+
"learning_rate": 4.356760282773209e-09,
|
| 2157 |
+
"loss": 0.2533,
|
| 2158 |
+
"step": 307
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.989558232931727,
|
| 2162 |
+
"grad_norm": 2.426079034805298,
|
| 2163 |
+
"learning_rate": 2.4508333966305473e-09,
|
| 2164 |
+
"loss": 0.3066,
|
| 2165 |
+
"step": 308
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.9927710843373494,
|
| 2169 |
+
"grad_norm": 2.5514280796051025,
|
| 2170 |
+
"learning_rate": 1.089308730043981e-09,
|
| 2171 |
+
"loss": 0.2522,
|
| 2172 |
+
"step": 309
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.9959839357429718,
|
| 2176 |
+
"grad_norm": 3.108274221420288,
|
| 2177 |
+
"learning_rate": 2.723345991245685e-10,
|
| 2178 |
+
"loss": 0.3181,
|
| 2179 |
+
"step": 310
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.9991967871485944,
|
| 2183 |
+
"grad_norm": 1.9873707294464111,
|
| 2184 |
+
"learning_rate": 0.0,
|
| 2185 |
+
"loss": 0.2276,
|
| 2186 |
+
"step": 311
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.9991967871485944,
|
| 2190 |
+
"step": 311,
|
| 2191 |
+
"total_flos": 9.654917969372774e+16,
|
| 2192 |
+
"train_loss": 0.38514137028497897,
|
| 2193 |
+
"train_runtime": 6718.4685,
|
| 2194 |
+
"train_samples_per_second": 1.483,
|
| 2195 |
+
"train_steps_per_second": 0.046
|
| 2196 |
+
}
|
| 2197 |
+
],
|
| 2198 |
+
"logging_steps": 1.0,
|
| 2199 |
+
"max_steps": 311,
|
| 2200 |
+
"num_input_tokens_seen": 0,
|
| 2201 |
+
"num_train_epochs": 1,
|
| 2202 |
+
"save_steps": 500,
|
| 2203 |
+
"stateful_callbacks": {
|
| 2204 |
+
"TrainerControl": {
|
| 2205 |
+
"args": {
|
| 2206 |
+
"should_epoch_stop": false,
|
| 2207 |
+
"should_evaluate": false,
|
| 2208 |
+
"should_log": false,
|
| 2209 |
+
"should_save": true,
|
| 2210 |
+
"should_training_stop": true
|
| 2211 |
+
},
|
| 2212 |
+
"attributes": {}
|
| 2213 |
+
}
|
| 2214 |
+
},
|
| 2215 |
+
"total_flos": 9.654917969372774e+16,
|
| 2216 |
+
"train_batch_size": 1,
|
| 2217 |
+
"trial_name": null,
|
| 2218 |
+
"trial_params": null
|
| 2219 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ba13fafd79b52a28f4ad9217a1e459b2364c55c4322937c73e141ed5407b98de
|
| 3 |
+
size 7352
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|