Model save
Browse files- README.md +65 -0
- all_results.json +9 -0
- generation_config.json +7 -0
- model.safetensors +1 -1
- runs/Jun30_22-33-16_poseidon/events.out.tfevents.1719786808.poseidon.1047944.0 +2 -2
- train_results.json +9 -0
- trainer_state.json +2122 -0
README.md
ADDED
|
@@ -0,0 +1,65 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T
|
| 4 |
+
tags:
|
| 5 |
+
- trl
|
| 6 |
+
- sft
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
datasets:
|
| 9 |
+
- generator
|
| 10 |
+
model-index:
|
| 11 |
+
- name: tinyllama-1.1b-sum-sft-full_LR4e-5
|
| 12 |
+
results: []
|
| 13 |
+
---
|
| 14 |
+
|
| 15 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 16 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 17 |
+
|
| 18 |
+
# tinyllama-1.1b-sum-sft-full_LR4e-5
|
| 19 |
+
|
| 20 |
+
This model is a fine-tuned version of [TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T](https://huggingface.co/TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T) on the generator dataset.
|
| 21 |
+
It achieves the following results on the evaluation set:
|
| 22 |
+
- Loss: 2.1087
|
| 23 |
+
|
| 24 |
+
## Model description
|
| 25 |
+
|
| 26 |
+
More information needed
|
| 27 |
+
|
| 28 |
+
## Intended uses & limitations
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Training and evaluation data
|
| 33 |
+
|
| 34 |
+
More information needed
|
| 35 |
+
|
| 36 |
+
## Training procedure
|
| 37 |
+
|
| 38 |
+
### Training hyperparameters
|
| 39 |
+
|
| 40 |
+
The following hyperparameters were used during training:
|
| 41 |
+
- learning_rate: 4e-05
|
| 42 |
+
- train_batch_size: 8
|
| 43 |
+
- eval_batch_size: 8
|
| 44 |
+
- seed: 42
|
| 45 |
+
- distributed_type: multi-GPU
|
| 46 |
+
- gradient_accumulation_steps: 2
|
| 47 |
+
- total_train_batch_size: 16
|
| 48 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 49 |
+
- lr_scheduler_type: cosine
|
| 50 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 51 |
+
- num_epochs: 1
|
| 52 |
+
|
| 53 |
+
### Training results
|
| 54 |
+
|
| 55 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 56 |
+
|:-------------:|:------:|:----:|:---------------:|
|
| 57 |
+
| 2.1044 | 0.9997 | 1476 | 2.1087 |
|
| 58 |
+
|
| 59 |
+
|
| 60 |
+
### Framework versions
|
| 61 |
+
|
| 62 |
+
- Transformers 4.41.2
|
| 63 |
+
- Pytorch 2.1.2
|
| 64 |
+
- Datasets 2.20.0
|
| 65 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9996613613274636,
|
| 3 |
+
"total_flos": 3.002086743311647e+17,
|
| 4 |
+
"train_loss": 2.1322910175090883,
|
| 5 |
+
"train_runtime": 6591.5154,
|
| 6 |
+
"train_samples": 116722,
|
| 7 |
+
"train_samples_per_second": 3.583,
|
| 8 |
+
"train_steps_per_second": 0.224
|
| 9 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 1,
|
| 3 |
+
"eos_token_id": 2,
|
| 4 |
+
"max_length": 2048,
|
| 5 |
+
"pad_token_id": 0,
|
| 6 |
+
"transformers_version": "4.41.2"
|
| 7 |
+
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2200119864
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:05ca16591df68398f82ff5ddec1e4efb72cf85d281518a1cc06998bcf4a2a7c8
|
| 3 |
size 2200119864
|
runs/Jun30_22-33-16_poseidon/events.out.tfevents.1719786808.poseidon.1047944.0
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be0e105153ef301392ab141122ee0cd4a7a189d119569bfed0e2541d0321deb7
|
| 3 |
+
size 68036
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9996613613274636,
|
| 3 |
+
"total_flos": 3.002086743311647e+17,
|
| 4 |
+
"train_loss": 2.1322910175090883,
|
| 5 |
+
"train_runtime": 6591.5154,
|
| 6 |
+
"train_samples": 116722,
|
| 7 |
+
"train_samples_per_second": 3.583,
|
| 8 |
+
"train_steps_per_second": 0.224
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2122 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9996613613274636,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1476,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0006772773450728074,
|
| 13 |
+
"grad_norm": 11.0625,
|
| 14 |
+
"learning_rate": 2.702702702702703e-07,
|
| 15 |
+
"loss": 2.4477,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0033863867253640365,
|
| 20 |
+
"grad_norm": 11.8125,
|
| 21 |
+
"learning_rate": 1.3513513513513515e-06,
|
| 22 |
+
"loss": 2.4737,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.006772773450728073,
|
| 27 |
+
"grad_norm": 11.75,
|
| 28 |
+
"learning_rate": 2.702702702702703e-06,
|
| 29 |
+
"loss": 2.457,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.01015916017609211,
|
| 34 |
+
"grad_norm": 5.46875,
|
| 35 |
+
"learning_rate": 4.0540540540540545e-06,
|
| 36 |
+
"loss": 2.4211,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.013545546901456146,
|
| 41 |
+
"grad_norm": 2.875,
|
| 42 |
+
"learning_rate": 5.405405405405406e-06,
|
| 43 |
+
"loss": 2.3635,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.016931933626820182,
|
| 48 |
+
"grad_norm": 2.109375,
|
| 49 |
+
"learning_rate": 6.7567567567567575e-06,
|
| 50 |
+
"loss": 2.3577,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.02031832035218422,
|
| 55 |
+
"grad_norm": 1.7265625,
|
| 56 |
+
"learning_rate": 8.108108108108109e-06,
|
| 57 |
+
"loss": 2.2991,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.023704707077548254,
|
| 62 |
+
"grad_norm": 1.671875,
|
| 63 |
+
"learning_rate": 9.45945945945946e-06,
|
| 64 |
+
"loss": 2.2859,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.027091093802912292,
|
| 69 |
+
"grad_norm": 1.7421875,
|
| 70 |
+
"learning_rate": 1.0810810810810812e-05,
|
| 71 |
+
"loss": 2.2701,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.03047748052827633,
|
| 76 |
+
"grad_norm": 1.375,
|
| 77 |
+
"learning_rate": 1.2162162162162164e-05,
|
| 78 |
+
"loss": 2.2274,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.033863867253640365,
|
| 83 |
+
"grad_norm": 1.3203125,
|
| 84 |
+
"learning_rate": 1.3513513513513515e-05,
|
| 85 |
+
"loss": 2.2163,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.0372502539790044,
|
| 90 |
+
"grad_norm": 1.3046875,
|
| 91 |
+
"learning_rate": 1.4864864864864865e-05,
|
| 92 |
+
"loss": 2.2257,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.04063664070436844,
|
| 97 |
+
"grad_norm": 1.3046875,
|
| 98 |
+
"learning_rate": 1.6216216216216218e-05,
|
| 99 |
+
"loss": 2.2233,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.04402302742973248,
|
| 104 |
+
"grad_norm": 1.28125,
|
| 105 |
+
"learning_rate": 1.756756756756757e-05,
|
| 106 |
+
"loss": 2.2121,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.04740941415509651,
|
| 111 |
+
"grad_norm": 1.2421875,
|
| 112 |
+
"learning_rate": 1.891891891891892e-05,
|
| 113 |
+
"loss": 2.1914,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.05079580088046055,
|
| 118 |
+
"grad_norm": 1.2734375,
|
| 119 |
+
"learning_rate": 2.0270270270270273e-05,
|
| 120 |
+
"loss": 2.2119,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.054182187605824585,
|
| 125 |
+
"grad_norm": 1.21875,
|
| 126 |
+
"learning_rate": 2.1621621621621624e-05,
|
| 127 |
+
"loss": 2.1958,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.05756857433118862,
|
| 132 |
+
"grad_norm": 1.28125,
|
| 133 |
+
"learning_rate": 2.2972972972972976e-05,
|
| 134 |
+
"loss": 2.1629,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.06095496105655266,
|
| 139 |
+
"grad_norm": 1.21875,
|
| 140 |
+
"learning_rate": 2.4324324324324327e-05,
|
| 141 |
+
"loss": 2.1922,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.06434134778191669,
|
| 146 |
+
"grad_norm": 1.21875,
|
| 147 |
+
"learning_rate": 2.567567567567568e-05,
|
| 148 |
+
"loss": 2.1729,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.06772773450728073,
|
| 153 |
+
"grad_norm": 1.3203125,
|
| 154 |
+
"learning_rate": 2.702702702702703e-05,
|
| 155 |
+
"loss": 2.1695,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.07111412123264477,
|
| 160 |
+
"grad_norm": 1.21875,
|
| 161 |
+
"learning_rate": 2.8378378378378378e-05,
|
| 162 |
+
"loss": 2.1656,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.0745005079580088,
|
| 167 |
+
"grad_norm": 1.265625,
|
| 168 |
+
"learning_rate": 2.972972972972973e-05,
|
| 169 |
+
"loss": 2.1775,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.07788689468337284,
|
| 174 |
+
"grad_norm": 1.265625,
|
| 175 |
+
"learning_rate": 3.108108108108108e-05,
|
| 176 |
+
"loss": 2.1819,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.08127328140873688,
|
| 181 |
+
"grad_norm": 1.2421875,
|
| 182 |
+
"learning_rate": 3.2432432432432436e-05,
|
| 183 |
+
"loss": 2.1626,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.08465966813410092,
|
| 188 |
+
"grad_norm": 1.1953125,
|
| 189 |
+
"learning_rate": 3.378378378378379e-05,
|
| 190 |
+
"loss": 2.1499,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.08804605485946496,
|
| 195 |
+
"grad_norm": 1.2734375,
|
| 196 |
+
"learning_rate": 3.513513513513514e-05,
|
| 197 |
+
"loss": 2.191,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.091432441584829,
|
| 202 |
+
"grad_norm": 1.234375,
|
| 203 |
+
"learning_rate": 3.648648648648649e-05,
|
| 204 |
+
"loss": 2.1903,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.09481882831019302,
|
| 209 |
+
"grad_norm": 1.265625,
|
| 210 |
+
"learning_rate": 3.783783783783784e-05,
|
| 211 |
+
"loss": 2.1469,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.09820521503555706,
|
| 216 |
+
"grad_norm": 1.2265625,
|
| 217 |
+
"learning_rate": 3.918918918918919e-05,
|
| 218 |
+
"loss": 2.1473,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.1015916017609211,
|
| 223 |
+
"grad_norm": 1.2109375,
|
| 224 |
+
"learning_rate": 3.9999776147073465e-05,
|
| 225 |
+
"loss": 2.148,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.10497798848628513,
|
| 230 |
+
"grad_norm": 1.234375,
|
| 231 |
+
"learning_rate": 3.999725785919791e-05,
|
| 232 |
+
"loss": 2.1608,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.10836437521164917,
|
| 237 |
+
"grad_norm": 1.2109375,
|
| 238 |
+
"learning_rate": 3.9991941820788456e-05,
|
| 239 |
+
"loss": 2.1793,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.11175076193701321,
|
| 244 |
+
"grad_norm": 1.2265625,
|
| 245 |
+
"learning_rate": 3.998382877559453e-05,
|
| 246 |
+
"loss": 2.1449,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.11513714866237724,
|
| 251 |
+
"grad_norm": 1.1953125,
|
| 252 |
+
"learning_rate": 3.99729198586856e-05,
|
| 253 |
+
"loss": 2.146,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.11852353538774128,
|
| 258 |
+
"grad_norm": 1.2109375,
|
| 259 |
+
"learning_rate": 3.995921659629232e-05,
|
| 260 |
+
"loss": 2.1624,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.12190992211310532,
|
| 265 |
+
"grad_norm": 1.2265625,
|
| 266 |
+
"learning_rate": 3.9942720905593045e-05,
|
| 267 |
+
"loss": 2.1611,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.12529630883846934,
|
| 272 |
+
"grad_norm": 1.171875,
|
| 273 |
+
"learning_rate": 3.992343509444555e-05,
|
| 274 |
+
"loss": 2.1585,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.12868269556383338,
|
| 279 |
+
"grad_norm": 1.2578125,
|
| 280 |
+
"learning_rate": 3.9901361861064215e-05,
|
| 281 |
+
"loss": 2.1458,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.13206908228919742,
|
| 286 |
+
"grad_norm": 1.2421875,
|
| 287 |
+
"learning_rate": 3.987650429364247e-05,
|
| 288 |
+
"loss": 2.1729,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.13545546901456146,
|
| 293 |
+
"grad_norm": 1.2109375,
|
| 294 |
+
"learning_rate": 3.984886586992077e-05,
|
| 295 |
+
"loss": 2.1515,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.1388418557399255,
|
| 300 |
+
"grad_norm": 1.140625,
|
| 301 |
+
"learning_rate": 3.9818450456700014e-05,
|
| 302 |
+
"loss": 2.1697,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.14222824246528953,
|
| 307 |
+
"grad_norm": 1.1796875,
|
| 308 |
+
"learning_rate": 3.978526230930056e-05,
|
| 309 |
+
"loss": 2.155,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.14561462919065357,
|
| 314 |
+
"grad_norm": 1.1796875,
|
| 315 |
+
"learning_rate": 3.97493060709669e-05,
|
| 316 |
+
"loss": 2.1506,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.1490010159160176,
|
| 321 |
+
"grad_norm": 1.15625,
|
| 322 |
+
"learning_rate": 3.971058677221799e-05,
|
| 323 |
+
"loss": 2.1427,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.15238740264138165,
|
| 328 |
+
"grad_norm": 1.2578125,
|
| 329 |
+
"learning_rate": 3.966910983014349e-05,
|
| 330 |
+
"loss": 2.1601,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.15577378936674569,
|
| 335 |
+
"grad_norm": 1.1953125,
|
| 336 |
+
"learning_rate": 3.962488104764586e-05,
|
| 337 |
+
"loss": 2.1503,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.15916017609210972,
|
| 342 |
+
"grad_norm": 1.171875,
|
| 343 |
+
"learning_rate": 3.95779066126285e-05,
|
| 344 |
+
"loss": 2.1404,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.16254656281747376,
|
| 349 |
+
"grad_norm": 1.1875,
|
| 350 |
+
"learning_rate": 3.952819309713002e-05,
|
| 351 |
+
"loss": 2.1468,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.1659329495428378,
|
| 356 |
+
"grad_norm": 1.1875,
|
| 357 |
+
"learning_rate": 3.947574745640475e-05,
|
| 358 |
+
"loss": 2.1482,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.16931933626820184,
|
| 363 |
+
"grad_norm": 1.171875,
|
| 364 |
+
"learning_rate": 3.942057702794969e-05,
|
| 365 |
+
"loss": 2.1544,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.17270572299356587,
|
| 370 |
+
"grad_norm": 1.1796875,
|
| 371 |
+
"learning_rate": 3.9362689530477915e-05,
|
| 372 |
+
"loss": 2.1575,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.1760921097189299,
|
| 377 |
+
"grad_norm": 1.1953125,
|
| 378 |
+
"learning_rate": 3.930209306283867e-05,
|
| 379 |
+
"loss": 2.1607,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.17947849644429395,
|
| 384 |
+
"grad_norm": 1.1796875,
|
| 385 |
+
"learning_rate": 3.923879610288432e-05,
|
| 386 |
+
"loss": 2.1627,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.182864883169658,
|
| 391 |
+
"grad_norm": 1.203125,
|
| 392 |
+
"learning_rate": 3.917280750628421e-05,
|
| 393 |
+
"loss": 2.1424,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.186251269895022,
|
| 398 |
+
"grad_norm": 1.1640625,
|
| 399 |
+
"learning_rate": 3.9104136505285716e-05,
|
| 400 |
+
"loss": 2.1607,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.18963765662038604,
|
| 405 |
+
"grad_norm": 1.1875,
|
| 406 |
+
"learning_rate": 3.903279270742259e-05,
|
| 407 |
+
"loss": 2.136,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.19302404334575007,
|
| 412 |
+
"grad_norm": 1.1953125,
|
| 413 |
+
"learning_rate": 3.8958786094170784e-05,
|
| 414 |
+
"loss": 2.1558,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.1964104300711141,
|
| 419 |
+
"grad_norm": 1.1953125,
|
| 420 |
+
"learning_rate": 3.8882127019552006e-05,
|
| 421 |
+
"loss": 2.1368,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.19979681679647815,
|
| 426 |
+
"grad_norm": 1.171875,
|
| 427 |
+
"learning_rate": 3.880282620868507e-05,
|
| 428 |
+
"loss": 2.1243,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.2031832035218422,
|
| 433 |
+
"grad_norm": 1.1796875,
|
| 434 |
+
"learning_rate": 3.872089475628546e-05,
|
| 435 |
+
"loss": 2.1518,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.20656959024720623,
|
| 440 |
+
"grad_norm": 1.15625,
|
| 441 |
+
"learning_rate": 3.863634412511302e-05,
|
| 442 |
+
"loss": 2.12,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.20995597697257026,
|
| 447 |
+
"grad_norm": 1.1484375,
|
| 448 |
+
"learning_rate": 3.8549186144368304e-05,
|
| 449 |
+
"loss": 2.172,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.2133423636979343,
|
| 454 |
+
"grad_norm": 1.15625,
|
| 455 |
+
"learning_rate": 3.845943300803754e-05,
|
| 456 |
+
"loss": 2.1216,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.21672875042329834,
|
| 461 |
+
"grad_norm": 1.140625,
|
| 462 |
+
"learning_rate": 3.8367097273186644e-05,
|
| 463 |
+
"loss": 2.1387,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.22011513714866238,
|
| 468 |
+
"grad_norm": 1.1875,
|
| 469 |
+
"learning_rate": 3.827219185820441e-05,
|
| 470 |
+
"loss": 2.1338,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.22350152387402641,
|
| 475 |
+
"grad_norm": 1.1796875,
|
| 476 |
+
"learning_rate": 3.817473004099508e-05,
|
| 477 |
+
"loss": 2.134,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.22688791059939045,
|
| 482 |
+
"grad_norm": 1.1640625,
|
| 483 |
+
"learning_rate": 3.807472545712076e-05,
|
| 484 |
+
"loss": 2.1619,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.2302742973247545,
|
| 489 |
+
"grad_norm": 1.1640625,
|
| 490 |
+
"learning_rate": 3.797219209789365e-05,
|
| 491 |
+
"loss": 2.129,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.23366068405011853,
|
| 496 |
+
"grad_norm": 1.1484375,
|
| 497 |
+
"learning_rate": 3.786714430841858e-05,
|
| 498 |
+
"loss": 2.1493,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.23704707077548257,
|
| 503 |
+
"grad_norm": 1.1796875,
|
| 504 |
+
"learning_rate": 3.7759596785586066e-05,
|
| 505 |
+
"loss": 2.1453,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.2404334575008466,
|
| 510 |
+
"grad_norm": 1.1328125,
|
| 511 |
+
"learning_rate": 3.7649564576016076e-05,
|
| 512 |
+
"loss": 2.1411,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.24381984422621064,
|
| 517 |
+
"grad_norm": 1.2109375,
|
| 518 |
+
"learning_rate": 3.7537063073952904e-05,
|
| 519 |
+
"loss": 2.1367,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.24720623095157468,
|
| 524 |
+
"grad_norm": 1.1953125,
|
| 525 |
+
"learning_rate": 3.742210801911147e-05,
|
| 526 |
+
"loss": 2.1542,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.2505926176769387,
|
| 531 |
+
"grad_norm": 1.2109375,
|
| 532 |
+
"learning_rate": 3.730471549447516e-05,
|
| 533 |
+
"loss": 2.1327,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.25397900440230275,
|
| 538 |
+
"grad_norm": 1.1328125,
|
| 539 |
+
"learning_rate": 3.7184901924045745e-05,
|
| 540 |
+
"loss": 2.1263,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.25736539112766676,
|
| 545 |
+
"grad_norm": 1.15625,
|
| 546 |
+
"learning_rate": 3.7062684070545535e-05,
|
| 547 |
+
"loss": 2.1543,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.26075177785303083,
|
| 552 |
+
"grad_norm": 1.2109375,
|
| 553 |
+
"learning_rate": 3.69380790330722e-05,
|
| 554 |
+
"loss": 2.1336,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.26413816457839484,
|
| 559 |
+
"grad_norm": 1.1796875,
|
| 560 |
+
"learning_rate": 3.6811104244706424e-05,
|
| 561 |
+
"loss": 2.148,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.2675245513037589,
|
| 566 |
+
"grad_norm": 1.203125,
|
| 567 |
+
"learning_rate": 3.668177747007297e-05,
|
| 568 |
+
"loss": 2.1161,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.2709109380291229,
|
| 573 |
+
"grad_norm": 1.1875,
|
| 574 |
+
"learning_rate": 3.6550116802855244e-05,
|
| 575 |
+
"loss": 2.1428,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.274297324754487,
|
| 580 |
+
"grad_norm": 1.171875,
|
| 581 |
+
"learning_rate": 3.6416140663263885e-05,
|
| 582 |
+
"loss": 2.1307,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.277683711479851,
|
| 587 |
+
"grad_norm": 1.1640625,
|
| 588 |
+
"learning_rate": 3.627986779545967e-05,
|
| 589 |
+
"loss": 2.1407,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.28107009820521506,
|
| 594 |
+
"grad_norm": 1.125,
|
| 595 |
+
"learning_rate": 3.614131726493102e-05,
|
| 596 |
+
"loss": 2.1198,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.28445648493057907,
|
| 601 |
+
"grad_norm": 1.171875,
|
| 602 |
+
"learning_rate": 3.600050845582669e-05,
|
| 603 |
+
"loss": 2.1339,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.28784287165594313,
|
| 608 |
+
"grad_norm": 1.09375,
|
| 609 |
+
"learning_rate": 3.5857461068243744e-05,
|
| 610 |
+
"loss": 2.1346,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.29122925838130714,
|
| 615 |
+
"grad_norm": 1.125,
|
| 616 |
+
"learning_rate": 3.5712195115471394e-05,
|
| 617 |
+
"loss": 2.1184,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.2946156451066712,
|
| 622 |
+
"grad_norm": 1.1953125,
|
| 623 |
+
"learning_rate": 3.5564730921191e-05,
|
| 624 |
+
"loss": 2.1343,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.2980020318320352,
|
| 629 |
+
"grad_norm": 1.1328125,
|
| 630 |
+
"learning_rate": 3.5415089116632665e-05,
|
| 631 |
+
"loss": 2.1285,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.30138841855739923,
|
| 636 |
+
"grad_norm": 1.171875,
|
| 637 |
+
"learning_rate": 3.526329063768878e-05,
|
| 638 |
+
"loss": 2.1209,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.3047748052827633,
|
| 643 |
+
"grad_norm": 1.125,
|
| 644 |
+
"learning_rate": 3.510935672198495e-05,
|
| 645 |
+
"loss": 2.1302,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.3081611920081273,
|
| 650 |
+
"grad_norm": 1.171875,
|
| 651 |
+
"learning_rate": 3.495330890590871e-05,
|
| 652 |
+
"loss": 2.1075,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.31154757873349137,
|
| 657 |
+
"grad_norm": 1.140625,
|
| 658 |
+
"learning_rate": 3.4795169021596417e-05,
|
| 659 |
+
"loss": 2.1173,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.3149339654588554,
|
| 664 |
+
"grad_norm": 1.1640625,
|
| 665 |
+
"learning_rate": 3.463495919387885e-05,
|
| 666 |
+
"loss": 2.1284,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.31832035218421945,
|
| 671 |
+
"grad_norm": 1.1640625,
|
| 672 |
+
"learning_rate": 3.447270183718574e-05,
|
| 673 |
+
"loss": 2.1303,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.32170673890958346,
|
| 678 |
+
"grad_norm": 1.1796875,
|
| 679 |
+
"learning_rate": 3.430841965240983e-05,
|
| 680 |
+
"loss": 2.1401,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.3250931256349475,
|
| 685 |
+
"grad_norm": 1.140625,
|
| 686 |
+
"learning_rate": 3.4142135623730954e-05,
|
| 687 |
+
"loss": 2.1344,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.32847951236031153,
|
| 692 |
+
"grad_norm": 1.140625,
|
| 693 |
+
"learning_rate": 3.397387301540028e-05,
|
| 694 |
+
"loss": 2.1269,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.3318658990856756,
|
| 699 |
+
"grad_norm": 1.1171875,
|
| 700 |
+
"learning_rate": 3.380365536848558e-05,
|
| 701 |
+
"loss": 2.1308,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.3352522858110396,
|
| 706 |
+
"grad_norm": 1.0859375,
|
| 707 |
+
"learning_rate": 3.363150649757763e-05,
|
| 708 |
+
"loss": 2.1456,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.3386386725364037,
|
| 713 |
+
"grad_norm": 1.1640625,
|
| 714 |
+
"learning_rate": 3.345745048745838e-05,
|
| 715 |
+
"loss": 2.1163,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.3420250592617677,
|
| 720 |
+
"grad_norm": 1.1328125,
|
| 721 |
+
"learning_rate": 3.328151168973139e-05,
|
| 722 |
+
"loss": 2.1115,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.34541144598713175,
|
| 727 |
+
"grad_norm": 1.1640625,
|
| 728 |
+
"learning_rate": 3.3103714719414814e-05,
|
| 729 |
+
"loss": 2.1203,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.34879783271249576,
|
| 734 |
+
"grad_norm": 1.15625,
|
| 735 |
+
"learning_rate": 3.2924084451497606e-05,
|
| 736 |
+
"loss": 2.1226,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.3521842194378598,
|
| 741 |
+
"grad_norm": 1.1328125,
|
| 742 |
+
"learning_rate": 3.274264601745938e-05,
|
| 743 |
+
"loss": 2.1006,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.35557060616322383,
|
| 748 |
+
"grad_norm": 1.140625,
|
| 749 |
+
"learning_rate": 3.25594248017543e-05,
|
| 750 |
+
"loss": 2.1031,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.3589569928885879,
|
| 755 |
+
"grad_norm": 1.171875,
|
| 756 |
+
"learning_rate": 3.237444643825965e-05,
|
| 757 |
+
"loss": 2.128,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.3623433796139519,
|
| 762 |
+
"grad_norm": 1.109375,
|
| 763 |
+
"learning_rate": 3.21877368066895e-05,
|
| 764 |
+
"loss": 2.1303,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.365729766339316,
|
| 769 |
+
"grad_norm": 1.1484375,
|
| 770 |
+
"learning_rate": 3.199932202897391e-05,
|
| 771 |
+
"loss": 2.1269,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.36911615306468,
|
| 776 |
+
"grad_norm": 1.140625,
|
| 777 |
+
"learning_rate": 3.1809228465604345e-05,
|
| 778 |
+
"loss": 2.1526,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.372502539790044,
|
| 783 |
+
"grad_norm": 1.1015625,
|
| 784 |
+
"learning_rate": 3.1617482711945616e-05,
|
| 785 |
+
"loss": 2.1275,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.37588892651540806,
|
| 790 |
+
"grad_norm": 1.109375,
|
| 791 |
+
"learning_rate": 3.1424111594515085e-05,
|
| 792 |
+
"loss": 2.1329,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.37927531324077207,
|
| 797 |
+
"grad_norm": 1.1328125,
|
| 798 |
+
"learning_rate": 3.1229142167229366e-05,
|
| 799 |
+
"loss": 2.1363,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.38266169996613614,
|
| 804 |
+
"grad_norm": 1.140625,
|
| 805 |
+
"learning_rate": 3.1032601707619375e-05,
|
| 806 |
+
"loss": 2.1249,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.38604808669150015,
|
| 811 |
+
"grad_norm": 1.125,
|
| 812 |
+
"learning_rate": 3.083451771301399e-05,
|
| 813 |
+
"loss": 2.1323,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.3894344734168642,
|
| 818 |
+
"grad_norm": 1.1171875,
|
| 819 |
+
"learning_rate": 3.0634917896692965e-05,
|
| 820 |
+
"loss": 2.1328,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.3928208601422282,
|
| 825 |
+
"grad_norm": 1.15625,
|
| 826 |
+
"learning_rate": 3.0433830184009694e-05,
|
| 827 |
+
"loss": 2.1177,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.3962072468675923,
|
| 832 |
+
"grad_norm": 1.1484375,
|
| 833 |
+
"learning_rate": 3.023128270848427e-05,
|
| 834 |
+
"loss": 2.1293,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.3995936335929563,
|
| 839 |
+
"grad_norm": 1.09375,
|
| 840 |
+
"learning_rate": 3.002730380786737e-05,
|
| 841 |
+
"loss": 2.1073,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.40298002031832036,
|
| 846 |
+
"grad_norm": 1.125,
|
| 847 |
+
"learning_rate": 2.982192202017568e-05,
|
| 848 |
+
"loss": 2.1186,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.4063664070436844,
|
| 853 |
+
"grad_norm": 1.140625,
|
| 854 |
+
"learning_rate": 2.9615166079699178e-05,
|
| 855 |
+
"loss": 2.1191,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.40975279376904844,
|
| 860 |
+
"grad_norm": 1.125,
|
| 861 |
+
"learning_rate": 2.9407064912981035e-05,
|
| 862 |
+
"loss": 2.1293,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.41313918049441245,
|
| 867 |
+
"grad_norm": 1.125,
|
| 868 |
+
"learning_rate": 2.91976476347706e-05,
|
| 869 |
+
"loss": 2.1106,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.4165255672197765,
|
| 874 |
+
"grad_norm": 1.109375,
|
| 875 |
+
"learning_rate": 2.898694354395006e-05,
|
| 876 |
+
"loss": 2.1094,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.4199119539451405,
|
| 881 |
+
"grad_norm": 1.1484375,
|
| 882 |
+
"learning_rate": 2.8774982119435307e-05,
|
| 883 |
+
"loss": 2.1175,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.4232983406705046,
|
| 888 |
+
"grad_norm": 1.1484375,
|
| 889 |
+
"learning_rate": 2.856179301605165e-05,
|
| 890 |
+
"loss": 2.1291,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.4266847273958686,
|
| 895 |
+
"grad_norm": 1.109375,
|
| 896 |
+
"learning_rate": 2.8347406060384934e-05,
|
| 897 |
+
"loss": 2.1246,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.43007111412123267,
|
| 902 |
+
"grad_norm": 1.1640625,
|
| 903 |
+
"learning_rate": 2.8131851246608533e-05,
|
| 904 |
+
"loss": 2.1263,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.4334575008465967,
|
| 909 |
+
"grad_norm": 1.1328125,
|
| 910 |
+
"learning_rate": 2.7915158732287042e-05,
|
| 911 |
+
"loss": 2.1355,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.43684388757196074,
|
| 916 |
+
"grad_norm": 1.109375,
|
| 917 |
+
"learning_rate": 2.7697358834156974e-05,
|
| 918 |
+
"loss": 2.124,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.44023027429732475,
|
| 923 |
+
"grad_norm": 1.1640625,
|
| 924 |
+
"learning_rate": 2.747848202388528e-05,
|
| 925 |
+
"loss": 2.107,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.4436166610226888,
|
| 930 |
+
"grad_norm": 1.1328125,
|
| 931 |
+
"learning_rate": 2.725855892380613e-05,
|
| 932 |
+
"loss": 2.1343,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.44700304774805283,
|
| 937 |
+
"grad_norm": 1.140625,
|
| 938 |
+
"learning_rate": 2.703762030263666e-05,
|
| 939 |
+
"loss": 2.1117,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.45038943447341684,
|
| 944 |
+
"grad_norm": 1.1015625,
|
| 945 |
+
"learning_rate": 2.6815697071172245e-05,
|
| 946 |
+
"loss": 2.1107,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.4537758211987809,
|
| 951 |
+
"grad_norm": 1.1640625,
|
| 952 |
+
"learning_rate": 2.6592820277961805e-05,
|
| 953 |
+
"loss": 2.1012,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.4571622079241449,
|
| 958 |
+
"grad_norm": 1.140625,
|
| 959 |
+
"learning_rate": 2.6369021104963976e-05,
|
| 960 |
+
"loss": 2.1291,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.460548594649509,
|
| 965 |
+
"grad_norm": 1.1484375,
|
| 966 |
+
"learning_rate": 2.61443308631845e-05,
|
| 967 |
+
"loss": 2.1062,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.463934981374873,
|
| 972 |
+
"grad_norm": 1.109375,
|
| 973 |
+
"learning_rate": 2.591878098829563e-05,
|
| 974 |
+
"loss": 2.1077,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.46732136810023706,
|
| 979 |
+
"grad_norm": 1.140625,
|
| 980 |
+
"learning_rate": 2.5692403036238038e-05,
|
| 981 |
+
"loss": 2.1231,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.47070775482560107,
|
| 986 |
+
"grad_norm": 1.1328125,
|
| 987 |
+
"learning_rate": 2.546522867880598e-05,
|
| 988 |
+
"loss": 2.1441,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.47409414155096513,
|
| 993 |
+
"grad_norm": 1.140625,
|
| 994 |
+
"learning_rate": 2.5237289699216136e-05,
|
| 995 |
+
"loss": 2.109,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.47748052827632914,
|
| 1000 |
+
"grad_norm": 1.140625,
|
| 1001 |
+
"learning_rate": 2.5008617987660975e-05,
|
| 1002 |
+
"loss": 2.1384,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.4808669150016932,
|
| 1007 |
+
"grad_norm": 1.125,
|
| 1008 |
+
"learning_rate": 2.4779245536847074e-05,
|
| 1009 |
+
"loss": 2.1083,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.4842533017270572,
|
| 1014 |
+
"grad_norm": 1.1328125,
|
| 1015 |
+
"learning_rate": 2.4549204437519137e-05,
|
| 1016 |
+
"loss": 2.1149,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.4876396884524213,
|
| 1021 |
+
"grad_norm": 1.09375,
|
| 1022 |
+
"learning_rate": 2.4318526873970278e-05,
|
| 1023 |
+
"loss": 2.1059,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.4910260751777853,
|
| 1028 |
+
"grad_norm": 1.125,
|
| 1029 |
+
"learning_rate": 2.4087245119539227e-05,
|
| 1030 |
+
"loss": 2.1218,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.49441246190314936,
|
| 1035 |
+
"grad_norm": 1.109375,
|
| 1036 |
+
"learning_rate": 2.385539153209508e-05,
|
| 1037 |
+
"loss": 2.112,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.49779884862851337,
|
| 1042 |
+
"grad_norm": 1.15625,
|
| 1043 |
+
"learning_rate": 2.3622998549510197e-05,
|
| 1044 |
+
"loss": 2.1128,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.5011852353538774,
|
| 1049 |
+
"grad_norm": 1.1328125,
|
| 1050 |
+
"learning_rate": 2.3390098685121938e-05,
|
| 1051 |
+
"loss": 2.1269,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.5045716220792414,
|
| 1056 |
+
"grad_norm": 1.1484375,
|
| 1057 |
+
"learning_rate": 2.3156724523183832e-05,
|
| 1058 |
+
"loss": 2.1233,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.5079580088046055,
|
| 1063 |
+
"grad_norm": 1.109375,
|
| 1064 |
+
"learning_rate": 2.2922908714306814e-05,
|
| 1065 |
+
"loss": 2.1443,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.5113443955299696,
|
| 1070 |
+
"grad_norm": 1.078125,
|
| 1071 |
+
"learning_rate": 2.268868397089119e-05,
|
| 1072 |
+
"loss": 2.0891,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.5147307822553335,
|
| 1077 |
+
"grad_norm": 1.1015625,
|
| 1078 |
+
"learning_rate": 2.2454083062549957e-05,
|
| 1079 |
+
"loss": 2.112,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.5181171689806976,
|
| 1084 |
+
"grad_norm": 1.1171875,
|
| 1085 |
+
"learning_rate": 2.2219138811524102e-05,
|
| 1086 |
+
"loss": 2.1152,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.5215035557060617,
|
| 1091 |
+
"grad_norm": 1.1328125,
|
| 1092 |
+
"learning_rate": 2.1983884088090553e-05,
|
| 1093 |
+
"loss": 2.1255,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.5248899424314256,
|
| 1098 |
+
"grad_norm": 1.109375,
|
| 1099 |
+
"learning_rate": 2.1748351805963422e-05,
|
| 1100 |
+
"loss": 2.1058,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.5282763291567897,
|
| 1105 |
+
"grad_norm": 1.125,
|
| 1106 |
+
"learning_rate": 2.151257491768914e-05,
|
| 1107 |
+
"loss": 2.1231,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.5316627158821537,
|
| 1112 |
+
"grad_norm": 1.109375,
|
| 1113 |
+
"learning_rate": 2.127658641003617e-05,
|
| 1114 |
+
"loss": 2.1258,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.5350491026075178,
|
| 1119 |
+
"grad_norm": 1.09375,
|
| 1120 |
+
"learning_rate": 2.1040419299379958e-05,
|
| 1121 |
+
"loss": 2.1146,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.5384354893328818,
|
| 1126 |
+
"grad_norm": 1.1171875,
|
| 1127 |
+
"learning_rate": 2.0804106627083708e-05,
|
| 1128 |
+
"loss": 2.1032,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.5418218760582458,
|
| 1133 |
+
"grad_norm": 1.1328125,
|
| 1134 |
+
"learning_rate": 2.0567681454875664e-05,
|
| 1135 |
+
"loss": 2.117,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.5452082627836099,
|
| 1140 |
+
"grad_norm": 1.125,
|
| 1141 |
+
"learning_rate": 2.0331176860223575e-05,
|
| 1142 |
+
"loss": 2.1131,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.548594649508974,
|
| 1147 |
+
"grad_norm": 1.09375,
|
| 1148 |
+
"learning_rate": 2.009462593170691e-05,
|
| 1149 |
+
"loss": 2.1134,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.5519810362343379,
|
| 1154 |
+
"grad_norm": 1.1328125,
|
| 1155 |
+
"learning_rate": 1.985806176438756e-05,
|
| 1156 |
+
"loss": 2.126,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.555367422959702,
|
| 1161 |
+
"grad_norm": 1.1171875,
|
| 1162 |
+
"learning_rate": 1.9621517455179627e-05,
|
| 1163 |
+
"loss": 2.1293,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.558753809685066,
|
| 1168 |
+
"grad_norm": 1.140625,
|
| 1169 |
+
"learning_rate": 1.93850260982189e-05,
|
| 1170 |
+
"loss": 2.1059,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.5621401964104301,
|
| 1175 |
+
"grad_norm": 1.15625,
|
| 1176 |
+
"learning_rate": 1.9148620780232842e-05,
|
| 1177 |
+
"loss": 2.1121,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.5655265831357941,
|
| 1182 |
+
"grad_norm": 1.1171875,
|
| 1183 |
+
"learning_rate": 1.891233457591142e-05,
|
| 1184 |
+
"loss": 2.1178,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.5689129698611581,
|
| 1189 |
+
"grad_norm": 1.109375,
|
| 1190 |
+
"learning_rate": 1.8676200543279864e-05,
|
| 1191 |
+
"loss": 2.105,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.5722993565865222,
|
| 1196 |
+
"grad_norm": 1.109375,
|
| 1197 |
+
"learning_rate": 1.8440251719073533e-05,
|
| 1198 |
+
"loss": 2.1097,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.5756857433118863,
|
| 1203 |
+
"grad_norm": 1.1015625,
|
| 1204 |
+
"learning_rate": 1.8204521114115872e-05,
|
| 1205 |
+
"loss": 2.1136,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.5790721300372502,
|
| 1210 |
+
"grad_norm": 1.1015625,
|
| 1211 |
+
"learning_rate": 1.7969041708700028e-05,
|
| 1212 |
+
"loss": 2.087,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.5824585167626143,
|
| 1217 |
+
"grad_norm": 1.109375,
|
| 1218 |
+
"learning_rate": 1.7733846447974606e-05,
|
| 1219 |
+
"loss": 2.115,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.5858449034879784,
|
| 1224 |
+
"grad_norm": 1.140625,
|
| 1225 |
+
"learning_rate": 1.7498968237334484e-05,
|
| 1226 |
+
"loss": 2.1076,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.5892312902133424,
|
| 1231 |
+
"grad_norm": 1.15625,
|
| 1232 |
+
"learning_rate": 1.7264439937817112e-05,
|
| 1233 |
+
"loss": 2.1182,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.5926176769387064,
|
| 1238 |
+
"grad_norm": 1.1171875,
|
| 1239 |
+
"learning_rate": 1.7030294361505007e-05,
|
| 1240 |
+
"loss": 2.1157,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.5960040636640704,
|
| 1245 |
+
"grad_norm": 1.125,
|
| 1246 |
+
"learning_rate": 1.6796564266935158e-05,
|
| 1247 |
+
"loss": 2.1072,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.5993904503894345,
|
| 1252 |
+
"grad_norm": 1.09375,
|
| 1253 |
+
"learning_rate": 1.6563282354515857e-05,
|
| 1254 |
+
"loss": 2.1034,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.6027768371147985,
|
| 1259 |
+
"grad_norm": 1.078125,
|
| 1260 |
+
"learning_rate": 1.6330481261951726e-05,
|
| 1261 |
+
"loss": 2.0906,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.6061632238401625,
|
| 1266 |
+
"grad_norm": 1.1328125,
|
| 1267 |
+
"learning_rate": 1.609819355967744e-05,
|
| 1268 |
+
"loss": 2.1431,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.6095496105655266,
|
| 1273 |
+
"grad_norm": 1.1015625,
|
| 1274 |
+
"learning_rate": 1.586645174630094e-05,
|
| 1275 |
+
"loss": 2.0905,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.6129359972908907,
|
| 1280 |
+
"grad_norm": 1.0859375,
|
| 1281 |
+
"learning_rate": 1.563528824405666e-05,
|
| 1282 |
+
"loss": 2.1059,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.6163223840162546,
|
| 1287 |
+
"grad_norm": 1.0703125,
|
| 1288 |
+
"learning_rate": 1.5404735394269403e-05,
|
| 1289 |
+
"loss": 2.1163,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.6197087707416187,
|
| 1294 |
+
"grad_norm": 1.0703125,
|
| 1295 |
+
"learning_rate": 1.5174825452829615e-05,
|
| 1296 |
+
"loss": 2.1211,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.6230951574669827,
|
| 1301 |
+
"grad_norm": 1.09375,
|
| 1302 |
+
"learning_rate": 1.4945590585680539e-05,
|
| 1303 |
+
"loss": 2.1069,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.6264815441923468,
|
| 1308 |
+
"grad_norm": 1.09375,
|
| 1309 |
+
"learning_rate": 1.4717062864318004e-05,
|
| 1310 |
+
"loss": 2.0974,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.6298679309177108,
|
| 1315 |
+
"grad_norm": 1.09375,
|
| 1316 |
+
"learning_rate": 1.4489274261303387e-05,
|
| 1317 |
+
"loss": 2.1034,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.6332543176430748,
|
| 1322 |
+
"grad_norm": 1.1015625,
|
| 1323 |
+
"learning_rate": 1.4262256645790469e-05,
|
| 1324 |
+
"loss": 2.1126,
|
| 1325 |
+
"step": 935
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.6366407043684389,
|
| 1329 |
+
"grad_norm": 1.109375,
|
| 1330 |
+
"learning_rate": 1.4036041779066696e-05,
|
| 1331 |
+
"loss": 2.1133,
|
| 1332 |
+
"step": 940
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.640027091093803,
|
| 1336 |
+
"grad_norm": 1.0625,
|
| 1337 |
+
"learning_rate": 1.3810661310109565e-05,
|
| 1338 |
+
"loss": 2.1165,
|
| 1339 |
+
"step": 945
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.6434134778191669,
|
| 1343 |
+
"grad_norm": 1.0859375,
|
| 1344 |
+
"learning_rate": 1.3586146771158746e-05,
|
| 1345 |
+
"loss": 2.1116,
|
| 1346 |
+
"step": 950
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.646799864544531,
|
| 1350 |
+
"grad_norm": 1.1328125,
|
| 1351 |
+
"learning_rate": 1.3362529573304473e-05,
|
| 1352 |
+
"loss": 2.0953,
|
| 1353 |
+
"step": 955
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.650186251269895,
|
| 1357 |
+
"grad_norm": 1.125,
|
| 1358 |
+
"learning_rate": 1.3139841002092949e-05,
|
| 1359 |
+
"loss": 2.1109,
|
| 1360 |
+
"step": 960
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.6535726379952591,
|
| 1364 |
+
"grad_norm": 1.109375,
|
| 1365 |
+
"learning_rate": 1.2918112213149292e-05,
|
| 1366 |
+
"loss": 2.0961,
|
| 1367 |
+
"step": 965
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.6569590247206231,
|
| 1371 |
+
"grad_norm": 1.1015625,
|
| 1372 |
+
"learning_rate": 1.2697374227818609e-05,
|
| 1373 |
+
"loss": 2.1,
|
| 1374 |
+
"step": 970
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.6603454114459871,
|
| 1378 |
+
"grad_norm": 1.1171875,
|
| 1379 |
+
"learning_rate": 1.2477657928825977e-05,
|
| 1380 |
+
"loss": 2.122,
|
| 1381 |
+
"step": 975
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.6637317981713512,
|
| 1385 |
+
"grad_norm": 1.125,
|
| 1386 |
+
"learning_rate": 1.2258994055955658e-05,
|
| 1387 |
+
"loss": 2.0896,
|
| 1388 |
+
"step": 980
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.6671181848967151,
|
| 1392 |
+
"grad_norm": 1.1015625,
|
| 1393 |
+
"learning_rate": 1.2041413201750473e-05,
|
| 1394 |
+
"loss": 2.0827,
|
| 1395 |
+
"step": 985
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.6705045716220792,
|
| 1399 |
+
"grad_norm": 1.0703125,
|
| 1400 |
+
"learning_rate": 1.1824945807231642e-05,
|
| 1401 |
+
"loss": 2.1213,
|
| 1402 |
+
"step": 990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.6738909583474433,
|
| 1406 |
+
"grad_norm": 1.0859375,
|
| 1407 |
+
"learning_rate": 1.1609622157639913e-05,
|
| 1408 |
+
"loss": 2.113,
|
| 1409 |
+
"step": 995
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.6772773450728073,
|
| 1413 |
+
"grad_norm": 1.125,
|
| 1414 |
+
"learning_rate": 1.139547237819846e-05,
|
| 1415 |
+
"loss": 2.1118,
|
| 1416 |
+
"step": 1000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.6806637317981713,
|
| 1420 |
+
"grad_norm": 1.109375,
|
| 1421 |
+
"learning_rate": 1.1182526429898118e-05,
|
| 1422 |
+
"loss": 2.1255,
|
| 1423 |
+
"step": 1005
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.6840501185235354,
|
| 1427 |
+
"grad_norm": 1.1328125,
|
| 1428 |
+
"learning_rate": 1.0970814105305689e-05,
|
| 1429 |
+
"loss": 2.116,
|
| 1430 |
+
"step": 1010
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.6874365052488994,
|
| 1434 |
+
"grad_norm": 1.1015625,
|
| 1435 |
+
"learning_rate": 1.0760365024395745e-05,
|
| 1436 |
+
"loss": 2.0958,
|
| 1437 |
+
"step": 1015
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.6908228919742635,
|
| 1441 |
+
"grad_norm": 1.1015625,
|
| 1442 |
+
"learning_rate": 1.0551208630406587e-05,
|
| 1443 |
+
"loss": 2.0891,
|
| 1444 |
+
"step": 1020
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.6942092786996275,
|
| 1448 |
+
"grad_norm": 1.1328125,
|
| 1449 |
+
"learning_rate": 1.0343374185720927e-05,
|
| 1450 |
+
"loss": 2.1199,
|
| 1451 |
+
"step": 1025
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.6975956654249915,
|
| 1455 |
+
"grad_norm": 1.109375,
|
| 1456 |
+
"learning_rate": 1.0136890767771923e-05,
|
| 1457 |
+
"loss": 2.1115,
|
| 1458 |
+
"step": 1030
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.7009820521503556,
|
| 1462 |
+
"grad_norm": 1.109375,
|
| 1463 |
+
"learning_rate": 9.931787264975021e-06,
|
| 1464 |
+
"loss": 2.1147,
|
| 1465 |
+
"step": 1035
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.7043684388757196,
|
| 1469 |
+
"grad_norm": 1.1015625,
|
| 1470 |
+
"learning_rate": 9.72809237268628e-06,
|
| 1471 |
+
"loss": 2.1142,
|
| 1472 |
+
"step": 1040
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.7077548256010836,
|
| 1476 |
+
"grad_norm": 1.09375,
|
| 1477 |
+
"learning_rate": 9.525834589187701e-06,
|
| 1478 |
+
"loss": 2.1012,
|
| 1479 |
+
"step": 1045
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.7111412123264477,
|
| 1483 |
+
"grad_norm": 1.1015625,
|
| 1484 |
+
"learning_rate": 9.325042211700111e-06,
|
| 1485 |
+
"loss": 2.1048,
|
| 1486 |
+
"step": 1050
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.7145275990518117,
|
| 1490 |
+
"grad_norm": 1.1171875,
|
| 1491 |
+
"learning_rate": 9.125743332424213e-06,
|
| 1492 |
+
"loss": 2.1197,
|
| 1493 |
+
"step": 1055
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.7179139857771758,
|
| 1497 |
+
"grad_norm": 1.078125,
|
| 1498 |
+
"learning_rate": 8.92796583461031e-06,
|
| 1499 |
+
"loss": 2.0983,
|
| 1500 |
+
"step": 1060
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.7213003725025398,
|
| 1504 |
+
"grad_norm": 1.1171875,
|
| 1505 |
+
"learning_rate": 8.731737388657198e-06,
|
| 1506 |
+
"loss": 2.1128,
|
| 1507 |
+
"step": 1065
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.7246867592279038,
|
| 1511 |
+
"grad_norm": 1.1015625,
|
| 1512 |
+
"learning_rate": 8.537085448240951e-06,
|
| 1513 |
+
"loss": 2.1188,
|
| 1514 |
+
"step": 1070
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.7280731459532679,
|
| 1518 |
+
"grad_norm": 1.1171875,
|
| 1519 |
+
"learning_rate": 8.344037246473919e-06,
|
| 1520 |
+
"loss": 2.1098,
|
| 1521 |
+
"step": 1075
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.731459532678632,
|
| 1525 |
+
"grad_norm": 1.1171875,
|
| 1526 |
+
"learning_rate": 8.152619792094674e-06,
|
| 1527 |
+
"loss": 2.1052,
|
| 1528 |
+
"step": 1080
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.7348459194039959,
|
| 1532 |
+
"grad_norm": 1.125,
|
| 1533 |
+
"learning_rate": 7.962859865689282e-06,
|
| 1534 |
+
"loss": 2.1407,
|
| 1535 |
+
"step": 1085
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.73823230612936,
|
| 1539 |
+
"grad_norm": 1.15625,
|
| 1540 |
+
"learning_rate": 7.77478401594453e-06,
|
| 1541 |
+
"loss": 2.1314,
|
| 1542 |
+
"step": 1090
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.741618692854724,
|
| 1546 |
+
"grad_norm": 1.1015625,
|
| 1547 |
+
"learning_rate": 7.588418555933581e-06,
|
| 1548 |
+
"loss": 2.101,
|
| 1549 |
+
"step": 1095
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.745005079580088,
|
| 1553 |
+
"grad_norm": 1.109375,
|
| 1554 |
+
"learning_rate": 7.403789559434573e-06,
|
| 1555 |
+
"loss": 2.1036,
|
| 1556 |
+
"step": 1100
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.7483914663054521,
|
| 1560 |
+
"grad_norm": 1.09375,
|
| 1561 |
+
"learning_rate": 7.220922857282804e-06,
|
| 1562 |
+
"loss": 2.0901,
|
| 1563 |
+
"step": 1105
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.7517778530308161,
|
| 1567 |
+
"grad_norm": 1.09375,
|
| 1568 |
+
"learning_rate": 7.039844033756713e-06,
|
| 1569 |
+
"loss": 2.088,
|
| 1570 |
+
"step": 1110
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.7551642397561802,
|
| 1574 |
+
"grad_norm": 1.140625,
|
| 1575 |
+
"learning_rate": 6.860578422998563e-06,
|
| 1576 |
+
"loss": 2.1146,
|
| 1577 |
+
"step": 1115
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.7585506264815441,
|
| 1581 |
+
"grad_norm": 1.1328125,
|
| 1582 |
+
"learning_rate": 6.683151105469956e-06,
|
| 1583 |
+
"loss": 2.1177,
|
| 1584 |
+
"step": 1120
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.7619370132069082,
|
| 1588 |
+
"grad_norm": 1.125,
|
| 1589 |
+
"learning_rate": 6.50758690444297e-06,
|
| 1590 |
+
"loss": 2.1109,
|
| 1591 |
+
"step": 1125
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.7653233999322723,
|
| 1595 |
+
"grad_norm": 1.1171875,
|
| 1596 |
+
"learning_rate": 6.333910382527175e-06,
|
| 1597 |
+
"loss": 2.1356,
|
| 1598 |
+
"step": 1130
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.7687097866576363,
|
| 1602 |
+
"grad_norm": 1.09375,
|
| 1603 |
+
"learning_rate": 6.1621458382331444e-06,
|
| 1604 |
+
"loss": 2.1003,
|
| 1605 |
+
"step": 1135
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.7720961733830003,
|
| 1609 |
+
"grad_norm": 1.1015625,
|
| 1610 |
+
"learning_rate": 5.9923173025729895e-06,
|
| 1611 |
+
"loss": 2.1078,
|
| 1612 |
+
"step": 1140
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.7754825601083644,
|
| 1616 |
+
"grad_norm": 1.140625,
|
| 1617 |
+
"learning_rate": 5.824448535698195e-06,
|
| 1618 |
+
"loss": 2.0915,
|
| 1619 |
+
"step": 1145
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.7788689468337284,
|
| 1623 |
+
"grad_norm": 1.140625,
|
| 1624 |
+
"learning_rate": 5.658563023575478e-06,
|
| 1625 |
+
"loss": 2.1269,
|
| 1626 |
+
"step": 1150
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.7822553335590925,
|
| 1630 |
+
"grad_norm": 1.09375,
|
| 1631 |
+
"learning_rate": 5.494683974700878e-06,
|
| 1632 |
+
"loss": 2.1221,
|
| 1633 |
+
"step": 1155
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.7856417202844564,
|
| 1637 |
+
"grad_norm": 1.1015625,
|
| 1638 |
+
"learning_rate": 5.332834316852786e-06,
|
| 1639 |
+
"loss": 2.0986,
|
| 1640 |
+
"step": 1160
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.7890281070098205,
|
| 1644 |
+
"grad_norm": 1.125,
|
| 1645 |
+
"learning_rate": 5.173036693884164e-06,
|
| 1646 |
+
"loss": 2.1099,
|
| 1647 |
+
"step": 1165
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.7924144937351846,
|
| 1651 |
+
"grad_norm": 1.1484375,
|
| 1652 |
+
"learning_rate": 5.015313462554527e-06,
|
| 1653 |
+
"loss": 2.1086,
|
| 1654 |
+
"step": 1170
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.7958008804605486,
|
| 1658 |
+
"grad_norm": 1.125,
|
| 1659 |
+
"learning_rate": 4.859686689402099e-06,
|
| 1660 |
+
"loss": 2.1064,
|
| 1661 |
+
"step": 1175
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.7991872671859126,
|
| 1665 |
+
"grad_norm": 1.125,
|
| 1666 |
+
"learning_rate": 4.70617814765651e-06,
|
| 1667 |
+
"loss": 2.1341,
|
| 1668 |
+
"step": 1180
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.8025736539112767,
|
| 1672 |
+
"grad_norm": 1.1015625,
|
| 1673 |
+
"learning_rate": 4.554809314192629e-06,
|
| 1674 |
+
"loss": 2.1181,
|
| 1675 |
+
"step": 1185
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.8059600406366407,
|
| 1679 |
+
"grad_norm": 1.1171875,
|
| 1680 |
+
"learning_rate": 4.405601366525776e-06,
|
| 1681 |
+
"loss": 2.1068,
|
| 1682 |
+
"step": 1190
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.8093464273620048,
|
| 1686 |
+
"grad_norm": 1.1328125,
|
| 1687 |
+
"learning_rate": 4.258575179848847e-06,
|
| 1688 |
+
"loss": 2.0982,
|
| 1689 |
+
"step": 1195
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.8127328140873687,
|
| 1693 |
+
"grad_norm": 1.09375,
|
| 1694 |
+
"learning_rate": 4.113751324111748e-06,
|
| 1695 |
+
"loss": 2.1218,
|
| 1696 |
+
"step": 1200
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.8161192008127328,
|
| 1700 |
+
"grad_norm": 1.0859375,
|
| 1701 |
+
"learning_rate": 3.971150061143492e-06,
|
| 1702 |
+
"loss": 2.1166,
|
| 1703 |
+
"step": 1205
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.8195055875380969,
|
| 1707 |
+
"grad_norm": 1.1015625,
|
| 1708 |
+
"learning_rate": 3.830791341817468e-06,
|
| 1709 |
+
"loss": 2.1201,
|
| 1710 |
+
"step": 1210
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.8228919742634608,
|
| 1714 |
+
"grad_norm": 1.109375,
|
| 1715 |
+
"learning_rate": 3.6926948032601663e-06,
|
| 1716 |
+
"loss": 2.1185,
|
| 1717 |
+
"step": 1215
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.8262783609888249,
|
| 1721 |
+
"grad_norm": 1.0703125,
|
| 1722 |
+
"learning_rate": 3.5568797661038004e-06,
|
| 1723 |
+
"loss": 2.1061,
|
| 1724 |
+
"step": 1220
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.829664747714189,
|
| 1728 |
+
"grad_norm": 1.125,
|
| 1729 |
+
"learning_rate": 3.4233652317832424e-06,
|
| 1730 |
+
"loss": 2.1188,
|
| 1731 |
+
"step": 1225
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.833051134439553,
|
| 1735 |
+
"grad_norm": 1.1015625,
|
| 1736 |
+
"learning_rate": 3.292169879877569e-06,
|
| 1737 |
+
"loss": 2.1077,
|
| 1738 |
+
"step": 1230
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.836437521164917,
|
| 1742 |
+
"grad_norm": 1.09375,
|
| 1743 |
+
"learning_rate": 3.1633120654966932e-06,
|
| 1744 |
+
"loss": 2.124,
|
| 1745 |
+
"step": 1235
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.839823907890281,
|
| 1749 |
+
"grad_norm": 1.109375,
|
| 1750 |
+
"learning_rate": 3.0368098167133376e-06,
|
| 1751 |
+
"loss": 2.1057,
|
| 1752 |
+
"step": 1240
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.8432102946156451,
|
| 1756 |
+
"grad_norm": 1.09375,
|
| 1757 |
+
"learning_rate": 2.9126808320408016e-06,
|
| 1758 |
+
"loss": 2.1011,
|
| 1759 |
+
"step": 1245
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.8465966813410092,
|
| 1763 |
+
"grad_norm": 1.09375,
|
| 1764 |
+
"learning_rate": 2.7909424779567996e-06,
|
| 1765 |
+
"loss": 2.0859,
|
| 1766 |
+
"step": 1250
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.8499830680663731,
|
| 1770 |
+
"grad_norm": 1.1015625,
|
| 1771 |
+
"learning_rate": 2.6716117864738024e-06,
|
| 1772 |
+
"loss": 2.1224,
|
| 1773 |
+
"step": 1255
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.8533694547917372,
|
| 1777 |
+
"grad_norm": 1.140625,
|
| 1778 |
+
"learning_rate": 2.5547054527561253e-06,
|
| 1779 |
+
"loss": 2.108,
|
| 1780 |
+
"step": 1260
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.8567558415171013,
|
| 1784 |
+
"grad_norm": 1.0859375,
|
| 1785 |
+
"learning_rate": 2.4402398327841658e-06,
|
| 1786 |
+
"loss": 2.1146,
|
| 1787 |
+
"step": 1265
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.8601422282424653,
|
| 1791 |
+
"grad_norm": 1.125,
|
| 1792 |
+
"learning_rate": 2.3282309410661007e-06,
|
| 1793 |
+
"loss": 2.1147,
|
| 1794 |
+
"step": 1270
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.8635286149678293,
|
| 1798 |
+
"grad_norm": 1.0859375,
|
| 1799 |
+
"learning_rate": 2.2186944483973273e-06,
|
| 1800 |
+
"loss": 2.087,
|
| 1801 |
+
"step": 1275
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.8669150016931934,
|
| 1805 |
+
"grad_norm": 1.125,
|
| 1806 |
+
"learning_rate": 2.1116456796680376e-06,
|
| 1807 |
+
"loss": 2.0946,
|
| 1808 |
+
"step": 1280
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.8703013884185574,
|
| 1812 |
+
"grad_norm": 1.09375,
|
| 1813 |
+
"learning_rate": 2.0070996117191676e-06,
|
| 1814 |
+
"loss": 2.115,
|
| 1815 |
+
"step": 1285
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.8736877751439215,
|
| 1819 |
+
"grad_norm": 1.125,
|
| 1820 |
+
"learning_rate": 1.9050708712470012e-06,
|
| 1821 |
+
"loss": 2.1077,
|
| 1822 |
+
"step": 1290
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.8770741618692854,
|
| 1826 |
+
"grad_norm": 1.1171875,
|
| 1827 |
+
"learning_rate": 1.8055737327568378e-06,
|
| 1828 |
+
"loss": 2.1076,
|
| 1829 |
+
"step": 1295
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.8804605485946495,
|
| 1833 |
+
"grad_norm": 1.0859375,
|
| 1834 |
+
"learning_rate": 1.7086221165658544e-06,
|
| 1835 |
+
"loss": 2.102,
|
| 1836 |
+
"step": 1300
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.8838469353200136,
|
| 1840 |
+
"grad_norm": 1.1171875,
|
| 1841 |
+
"learning_rate": 1.614229586855609e-06,
|
| 1842 |
+
"loss": 2.1115,
|
| 1843 |
+
"step": 1305
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.8872333220453776,
|
| 1847 |
+
"grad_norm": 1.09375,
|
| 1848 |
+
"learning_rate": 1.5224093497742654e-06,
|
| 1849 |
+
"loss": 2.1187,
|
| 1850 |
+
"step": 1310
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.8906197087707416,
|
| 1854 |
+
"grad_norm": 1.1015625,
|
| 1855 |
+
"learning_rate": 1.4331742515890091e-06,
|
| 1856 |
+
"loss": 2.1099,
|
| 1857 |
+
"step": 1315
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.8940060954961057,
|
| 1861 |
+
"grad_norm": 1.078125,
|
| 1862 |
+
"learning_rate": 1.3465367768887471e-06,
|
| 1863 |
+
"loss": 2.11,
|
| 1864 |
+
"step": 1320
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.8973924822214697,
|
| 1868 |
+
"grad_norm": 1.0625,
|
| 1869 |
+
"learning_rate": 1.2625090468374212e-06,
|
| 1870 |
+
"loss": 2.1088,
|
| 1871 |
+
"step": 1325
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.9007788689468337,
|
| 1875 |
+
"grad_norm": 1.1171875,
|
| 1876 |
+
"learning_rate": 1.1811028174782102e-06,
|
| 1877 |
+
"loss": 2.1239,
|
| 1878 |
+
"step": 1330
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.9041652556721977,
|
| 1882 |
+
"grad_norm": 1.078125,
|
| 1883 |
+
"learning_rate": 1.1023294780887394e-06,
|
| 1884 |
+
"loss": 2.1018,
|
| 1885 |
+
"step": 1335
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.9075516423975618,
|
| 1889 |
+
"grad_norm": 1.1328125,
|
| 1890 |
+
"learning_rate": 1.0262000495876735e-06,
|
| 1891 |
+
"loss": 2.12,
|
| 1892 |
+
"step": 1340
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.9109380291229259,
|
| 1896 |
+
"grad_norm": 1.09375,
|
| 1897 |
+
"learning_rate": 9.527251829927997e-07,
|
| 1898 |
+
"loss": 2.1071,
|
| 1899 |
+
"step": 1345
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.9143244158482898,
|
| 1903 |
+
"grad_norm": 1.09375,
|
| 1904 |
+
"learning_rate": 8.819151579308949e-07,
|
| 1905 |
+
"loss": 2.1099,
|
| 1906 |
+
"step": 1350
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.9177108025736539,
|
| 1910 |
+
"grad_norm": 1.109375,
|
| 1911 |
+
"learning_rate": 8.137798811995235e-07,
|
| 1912 |
+
"loss": 2.1087,
|
| 1913 |
+
"step": 1355
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.921097189299018,
|
| 1917 |
+
"grad_norm": 1.109375,
|
| 1918 |
+
"learning_rate": 7.483288853810067e-07,
|
| 1919 |
+
"loss": 2.1193,
|
| 1920 |
+
"step": 1360
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.924483576024382,
|
| 1924 |
+
"grad_norm": 1.1171875,
|
| 1925 |
+
"learning_rate": 6.85571327508765e-07,
|
| 1926 |
+
"loss": 2.1228,
|
| 1927 |
+
"step": 1365
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.927869962749746,
|
| 1931 |
+
"grad_norm": 1.1015625,
|
| 1932 |
+
"learning_rate": 6.255159877861782e-07,
|
| 1933 |
+
"loss": 2.1164,
|
| 1934 |
+
"step": 1370
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.93125634947511,
|
| 1938 |
+
"grad_norm": 1.109375,
|
| 1939 |
+
"learning_rate": 5.681712683581775e-07,
|
| 1940 |
+
"loss": 2.1116,
|
| 1941 |
+
"step": 1375
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.9346427362004741,
|
| 1945 |
+
"grad_norm": 1.1015625,
|
| 1946 |
+
"learning_rate": 5.135451921357337e-07,
|
| 1947 |
+
"loss": 2.0996,
|
| 1948 |
+
"step": 1380
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.9380291229258382,
|
| 1952 |
+
"grad_norm": 1.140625,
|
| 1953 |
+
"learning_rate": 4.616454016733851e-07,
|
| 1954 |
+
"loss": 2.1326,
|
| 1955 |
+
"step": 1385
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.9414155096512021,
|
| 1959 |
+
"grad_norm": 1.1171875,
|
| 1960 |
+
"learning_rate": 4.1247915810001205e-07,
|
| 1961 |
+
"loss": 2.1102,
|
| 1962 |
+
"step": 1390
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.9448018963765662,
|
| 1966 |
+
"grad_norm": 1.0546875,
|
| 1967 |
+
"learning_rate": 3.660533401029387e-07,
|
| 1968 |
+
"loss": 2.1122,
|
| 1969 |
+
"step": 1395
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.9481882831019303,
|
| 1973 |
+
"grad_norm": 1.1484375,
|
| 1974 |
+
"learning_rate": 3.2237444296557173e-07,
|
| 1975 |
+
"loss": 2.1179,
|
| 1976 |
+
"step": 1400
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.9515746698272943,
|
| 1980 |
+
"grad_norm": 1.125,
|
| 1981 |
+
"learning_rate": 2.81448577658654e-07,
|
| 1982 |
+
"loss": 2.1075,
|
| 1983 |
+
"step": 1405
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.9549610565526583,
|
| 1987 |
+
"grad_norm": 1.1015625,
|
| 1988 |
+
"learning_rate": 2.432814699853081e-07,
|
| 1989 |
+
"loss": 2.1126,
|
| 1990 |
+
"step": 1410
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.9583474432780223,
|
| 1994 |
+
"grad_norm": 1.0859375,
|
| 1995 |
+
"learning_rate": 2.078784597799599e-07,
|
| 1996 |
+
"loss": 2.106,
|
| 1997 |
+
"step": 1415
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.9617338300033864,
|
| 2001 |
+
"grad_norm": 1.109375,
|
| 2002 |
+
"learning_rate": 1.752445001612535e-07,
|
| 2003 |
+
"loss": 2.1234,
|
| 2004 |
+
"step": 1420
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.9651202167287504,
|
| 2008 |
+
"grad_norm": 1.09375,
|
| 2009 |
+
"learning_rate": 1.4538415683907236e-07,
|
| 2010 |
+
"loss": 2.0946,
|
| 2011 |
+
"step": 1425
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.9685066034541144,
|
| 2015 |
+
"grad_norm": 1.1171875,
|
| 2016 |
+
"learning_rate": 1.1830160747577924e-07,
|
| 2017 |
+
"loss": 2.0985,
|
| 2018 |
+
"step": 1430
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.9718929901794785,
|
| 2022 |
+
"grad_norm": 1.1953125,
|
| 2023 |
+
"learning_rate": 9.400064110172358e-08,
|
| 2024 |
+
"loss": 2.1185,
|
| 2025 |
+
"step": 1435
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.9752793769048426,
|
| 2029 |
+
"grad_norm": 1.109375,
|
| 2030 |
+
"learning_rate": 7.248465758513457e-08,
|
| 2031 |
+
"loss": 2.0938,
|
| 2032 |
+
"step": 1440
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.9786657636302065,
|
| 2036 |
+
"grad_norm": 1.09375,
|
| 2037 |
+
"learning_rate": 5.375666715645489e-08,
|
| 2038 |
+
"loss": 2.109,
|
| 2039 |
+
"step": 1445
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.9820521503555706,
|
| 2043 |
+
"grad_norm": 1.1015625,
|
| 2044 |
+
"learning_rate": 3.781928998718654e-08,
|
| 2045 |
+
"loss": 2.1146,
|
| 2046 |
+
"step": 1450
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.9854385370809347,
|
| 2050 |
+
"grad_norm": 1.09375,
|
| 2051 |
+
"learning_rate": 2.4674755823321794e-08,
|
| 2052 |
+
"loss": 2.1029,
|
| 2053 |
+
"step": 1455
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.9888249238062987,
|
| 2057 |
+
"grad_norm": 1.1015625,
|
| 2058 |
+
"learning_rate": 1.4324903673370583e-08,
|
| 2059 |
+
"loss": 2.1195,
|
| 2060 |
+
"step": 1460
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.9922113105316627,
|
| 2064 |
+
"grad_norm": 1.0859375,
|
| 2065 |
+
"learning_rate": 6.771181551088468e-09,
|
| 2066 |
+
"loss": 2.1055,
|
| 2067 |
+
"step": 1465
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.9955976972570267,
|
| 2071 |
+
"grad_norm": 1.125,
|
| 2072 |
+
"learning_rate": 2.014646272876508e-09,
|
| 2073 |
+
"loss": 2.1176,
|
| 2074 |
+
"step": 1470
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.9989840839823908,
|
| 2078 |
+
"grad_norm": 1.078125,
|
| 2079 |
+
"learning_rate": 5.596330993284937e-11,
|
| 2080 |
+
"loss": 2.1044,
|
| 2081 |
+
"step": 1475
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.9996613613274636,
|
| 2085 |
+
"eval_loss": 2.1087405681610107,
|
| 2086 |
+
"eval_runtime": 85.1679,
|
| 2087 |
+
"eval_samples_per_second": 15.323,
|
| 2088 |
+
"eval_steps_per_second": 1.926,
|
| 2089 |
+
"step": 1476
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 0.9996613613274636,
|
| 2093 |
+
"step": 1476,
|
| 2094 |
+
"total_flos": 3.002086743311647e+17,
|
| 2095 |
+
"train_loss": 2.1322910175090883,
|
| 2096 |
+
"train_runtime": 6591.5154,
|
| 2097 |
+
"train_samples_per_second": 3.583,
|
| 2098 |
+
"train_steps_per_second": 0.224
|
| 2099 |
+
}
|
| 2100 |
+
],
|
| 2101 |
+
"logging_steps": 5,
|
| 2102 |
+
"max_steps": 1476,
|
| 2103 |
+
"num_input_tokens_seen": 0,
|
| 2104 |
+
"num_train_epochs": 1,
|
| 2105 |
+
"save_steps": 100,
|
| 2106 |
+
"stateful_callbacks": {
|
| 2107 |
+
"TrainerControl": {
|
| 2108 |
+
"args": {
|
| 2109 |
+
"should_epoch_stop": false,
|
| 2110 |
+
"should_evaluate": false,
|
| 2111 |
+
"should_log": false,
|
| 2112 |
+
"should_save": true,
|
| 2113 |
+
"should_training_stop": false
|
| 2114 |
+
},
|
| 2115 |
+
"attributes": {}
|
| 2116 |
+
}
|
| 2117 |
+
},
|
| 2118 |
+
"total_flos": 3.002086743311647e+17,
|
| 2119 |
+
"train_batch_size": 8,
|
| 2120 |
+
"trial_name": null,
|
| 2121 |
+
"trial_params": null
|
| 2122 |
+
}
|