End of training

README.md

@@ -2,10 +2,16 @@
 license: apache-2.0
 base_model: mistralai/Mistral-7B-v0.1
 tags:
+- alignment-handbook
+- trl
+- orpo
+- generated_from_trainer
 - trl
 - orpo
 - alignment-handbook
 - generated_from_trainer
+datasets:
+- HuggingFaceH4/ultrafeedback_binarized
 model-index:
 - name: zephyr-7b-sft-full-orpo
   results: []
@@ -17,20 +23,20 @@ should probably proofread and complete it, then remove this comment. -->
 [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/statking/huggingface/runs/ehjj41t1)
 # zephyr-7b-sft-full-orpo
 
-This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on an unknown dataset.
+This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the HuggingFaceH4/ultrafeedback_binarized dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5088
-- Rewards/chosen: -0.0404
+- Loss: 0.5072
+- Rewards/chosen: -0.0401
 - Rewards/rejected: -0.0510
-- Rewards/accuracies: 0.6290
-- Rewards/margins: 0.0106
-- Logps/rejected: -1.0202
-- Logps/chosen: -0.8085
-- Logits/rejected: -2.5337
-- Logits/chosen: -2.5634
-- Nll Loss: 0.4741
-- Log Odds Ratio: -0.6379
-- Log Odds Chosen: 0.3305
+- Rewards/accuracies: 0.6230
+- Rewards/margins: 0.0109
+- Logps/rejected: -1.0200
+- Logps/chosen: -0.8015
+- Logits/rejected: -2.4533
+- Logits/chosen: -2.4851
+- Nll Loss: 0.4727
+- Log Odds Ratio: -0.6343
+- Log Odds Chosen: 0.3605
 
 ## Model description
 

all_results.json

@@ -1,5 +1,21 @@
 {
     "epoch": 0.9994756161510225,
+    "eval_log_odds_chosen": 0.36045119166374207,
+    "eval_log_odds_ratio": -0.6342776417732239,
+    "eval_logits/chosen": -2.4851083755493164,
+    "eval_logits/rejected": -2.4533143043518066,
+    "eval_logps/chosen": -0.8014506697654724,
+    "eval_logps/rejected": -1.020006775856018,
+    "eval_loss": 0.5071792006492615,
+    "eval_nll_loss": 0.4726640582084656,
+    "eval_rewards/accuracies": 0.6230158805847168,
+    "eval_rewards/chosen": -0.04007253423333168,
+    "eval_rewards/margins": 0.010927810333669186,
+    "eval_rewards/rejected": -0.05100034177303314,
+    "eval_runtime": 136.8627,
+    "eval_samples": 1994,
+    "eval_samples_per_second": 14.569,
+    "eval_steps_per_second": 0.46,
     "total_flos": 0.0,
     "train_loss": 0.5642813587989287,
     "train_runtime": 20357.789,

config.json

@@ -21,6 +21,6 @@
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.41.0.dev0",
-  "use_cache": false,
+  "use_cache": true,
   "vocab_size": 32000
 }

eval_results.json

@@ -0,0 +1,19 @@
+{
+    "epoch": 0.9994756161510225,
+    "eval_log_odds_chosen": 0.36045119166374207,
+    "eval_log_odds_ratio": -0.6342776417732239,
+    "eval_logits/chosen": -2.4851083755493164,
+    "eval_logits/rejected": -2.4533143043518066,
+    "eval_logps/chosen": -0.8014506697654724,
+    "eval_logps/rejected": -1.020006775856018,
+    "eval_loss": 0.5071792006492615,
+    "eval_nll_loss": 0.4726640582084656,
+    "eval_rewards/accuracies": 0.6230158805847168,
+    "eval_rewards/chosen": -0.04007253423333168,
+    "eval_rewards/margins": 0.010927810333669186,
+    "eval_rewards/rejected": -0.05100034177303314,
+    "eval_runtime": 136.8627,
+    "eval_samples": 1994,
+    "eval_samples_per_second": 14.569,
+    "eval_steps_per_second": 0.46
+}