End of training

README.md

@@ -2,9 +2,15 @@
 license: apache-2.0
 base_model: mistralai/Mistral-7B-v0.1
 tags:
+- alignment-handbook
 - trl
 - orpo
 - generated_from_trainer
+- trl
+- orpo
+- generated_from_trainer
+datasets:
+- argilla/dpo-mix-7k
 model-index:
 - name: mistral-7b-orpo-alignment-handbook
   results: []
@@ -15,7 +21,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # mistral-7b-orpo-alignment-handbook
 
-This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the None dataset.
+This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the argilla/dpo-mix-7k dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.8562
 - Rewards/chosen: -0.0394
@@ -23,12 +29,12 @@ It achieves the following results on the evaluation set:
 - Rewards/accuracies: 0.6615
 - Rewards/margins: 0.0091
 - Logps/rejected: -0.9709
-- Logps/chosen: -0.7880
-- Logits/rejected: -2.9507
-- Logits/chosen: -2.9399
+- Logps/chosen: -0.7882
+- Logits/rejected: -2.9442
+- Logits/chosen: -2.9335
 - Nll Loss: 0.8317
-- Log Odds Ratio: -0.6238
-- Log Odds Chosen: 0.3606
+- Log Odds Ratio: -0.6241
+- Log Odds Chosen: 0.3600
 
 ## Model description
 

all_results.json

@@ -1,5 +1,21 @@
 {
     "epoch": 2.99,
+    "eval_log_odds_chosen": 0.3599511682987213,
+    "eval_log_odds_ratio": -0.6241087913513184,
+    "eval_logits/chosen": -2.9335391521453857,
+    "eval_logits/rejected": -2.9442245960235596,
+    "eval_logps/chosen": -0.7882003784179688,
+    "eval_logps/rejected": -0.9709375500679016,
+    "eval_loss": 0.8561599850654602,
+    "eval_nll_loss": 0.8317307829856873,
+    "eval_rewards/accuracies": 0.6614583134651184,
+    "eval_rewards/chosen": -0.03941002115607262,
+    "eval_rewards/margins": 0.009136858396232128,
+    "eval_rewards/rejected": -0.04854688048362732,
+    "eval_runtime": 32.31,
+    "eval_samples": 750,
+    "eval_samples_per_second": 23.213,
+    "eval_steps_per_second": 0.743,
     "train_loss": 0.8898289559379456,
     "train_runtime": 3406.8155,
     "train_samples": 6750,

config.json

@@ -22,6 +22,6 @@
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.39.0.dev0",
-  "use_cache": false,
+  "use_cache": true,
   "vocab_size": 32002
 }

eval_results.json

@@ -0,0 +1,19 @@
+{
+    "epoch": 2.99,
+    "eval_log_odds_chosen": 0.3599511682987213,
+    "eval_log_odds_ratio": -0.6241087913513184,
+    "eval_logits/chosen": -2.9335391521453857,
+    "eval_logits/rejected": -2.9442245960235596,
+    "eval_logps/chosen": -0.7882003784179688,
+    "eval_logps/rejected": -0.9709375500679016,
+    "eval_loss": 0.8561599850654602,
+    "eval_nll_loss": 0.8317307829856873,
+    "eval_rewards/accuracies": 0.6614583134651184,
+    "eval_rewards/chosen": -0.03941002115607262,
+    "eval_rewards/margins": 0.009136858396232128,
+    "eval_rewards/rejected": -0.04854688048362732,
+    "eval_runtime": 32.31,
+    "eval_samples": 750,
+    "eval_samples_per_second": 23.213,
+    "eval_steps_per_second": 0.743
+}

runs/Mar27_08-37-28_alvarobartt-sharp-bobcat-1-0/events.out.tfevents.1711546821.alvarobartt-sharp-bobcat-1-0.498.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:99afbd8e0e342abaee64e8665263d1dc02df3bd581d0f9fe7b8268fc2a84eca2
+size 997