End of training

README.md

@@ -2,9 +2,15 @@
 license: llama3
 base_model: Magpie-Align/Llama-3-8B-Magpie-Pro-MT-SFT-v0.1
 tags:
+- alignment-handbook
 - trl
 - dpo
 - generated_from_trainer
+- trl
+- dpo
+- generated_from_trainer
+datasets:
+- princeton-nlp/llama3-ultrafeedback
 model-index:
 - name: Llama-3-8B-Magpie-Pro-MT-UltraDPO2
   results: []
@@ -15,17 +21,17 @@ should probably proofread and complete it, then remove this comment. -->
 
 # Llama-3-8B-Magpie-Pro-MT-UltraDPO2
 
-This model is a fine-tuned version of [Magpie-Align/Llama-3-8B-Magpie-Pro-MT-SFT-v0.1](https://huggingface.co/Magpie-Align/Llama-3-8B-Magpie-Pro-MT-SFT-v0.1) on an unknown dataset.
+This model is a fine-tuned version of [Magpie-Align/Llama-3-8B-Magpie-Pro-MT-SFT-v0.1](https://huggingface.co/Magpie-Align/Llama-3-8B-Magpie-Pro-MT-SFT-v0.1) on the princeton-nlp/llama3-ultrafeedback dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6088
-- Rewards/chosen: -1.6321
-- Rewards/rejected: -1.9785
-- Rewards/accuracies: 0.6829
-- Rewards/margins: 0.3464
-- Logps/rejected: -459.1051
-- Logps/chosen: -418.7560
-- Logits/rejected: -0.6440
-- Logits/chosen: -0.6435
+- Loss: 0.6084
+- Rewards/chosen: -1.6265
+- Rewards/rejected: -1.9735
+- Rewards/accuracies: 0.6809
+- Rewards/margins: 0.3470
+- Logps/rejected: -458.6070
+- Logps/chosen: -418.2021
+- Logits/rejected: -0.6447
+- Logits/chosen: -0.6439
 
 ## Model description
 

all_results.json

@@ -1,5 +1,18 @@
 {
     "epoch": 0.9982631930527722,
+    "eval_logits/chosen": -0.6438767313957214,
+    "eval_logits/rejected": -0.6447061896324158,
+    "eval_logps/chosen": -418.2021179199219,
+    "eval_logps/rejected": -458.6069641113281,
+    "eval_loss": 0.6083793044090271,
+    "eval_rewards/accuracies": 0.6808943152427673,
+    "eval_rewards/chosen": -1.6265408992767334,
+    "eval_rewards/margins": 0.346989244222641,
+    "eval_rewards/rejected": -1.9735301733016968,
+    "eval_runtime": 375.4309,
+    "eval_samples": 1961,
+    "eval_samples_per_second": 5.223,
+    "eval_steps_per_second": 0.328,
     "total_flos": 0.0,
     "train_loss": 0.6321173631915189,
     "train_runtime": 21471.9268,

config.json

@@ -24,6 +24,6 @@
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.41.2",
-  "use_cache": false,
+  "use_cache": true,
   "vocab_size": 128256
 }

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 0.9982631930527722,
+    "eval_logits/chosen": -0.6438767313957214,
+    "eval_logits/rejected": -0.6447061896324158,
+    "eval_logps/chosen": -418.2021179199219,
+    "eval_logps/rejected": -458.6069641113281,
+    "eval_loss": 0.6083793044090271,
+    "eval_rewards/accuracies": 0.6808943152427673,
+    "eval_rewards/chosen": -1.6265408992767334,
+    "eval_rewards/margins": 0.346989244222641,
+    "eval_rewards/rejected": -1.9735301733016968,
+    "eval_runtime": 375.4309,
+    "eval_samples": 1961,
+    "eval_samples_per_second": 5.223,
+    "eval_steps_per_second": 0.328
+}

runs/Jun29_16-07-00_a1002/events.out.tfevents.1719699823.a1002.181973.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:637d27a18cad41c444df2e8f7d6213069de91f5a7bf09c224f76118aa1f02514
+size 828