End of training

README.md

@@ -3,9 +3,16 @@ library_name: transformers
 license: apache-2.0
 base_model: TinyLlama/TinyLlama-1.1B-Chat-v1.0
 tags:
+- alignment-handbook
 - trl
 - dpo
 - generated_from_trainer
+- trl
+- dpo
+- generated_from_trainer
+datasets:
+- data/ui_math
+- data/ui_coding
 model-index:
 - name: tinyllama-1.1b-chat-v1.0-ui-math-coding-dpo
   results: []
@@ -16,7 +23,17 @@ should probably proofread and complete it, then remove this comment. -->
 
 # tinyllama-1.1b-chat-v1.0-ui-math-coding-dpo
 
-This model is a fine-tuned version of [TinyLlama/TinyLlama-1.1B-Chat-v1.0](https://huggingface.co/TinyLlama/TinyLlama-1.1B-Chat-v1.0) on the None dataset.
+This model is a fine-tuned version of [TinyLlama/TinyLlama-1.1B-Chat-v1.0](https://huggingface.co/TinyLlama/TinyLlama-1.1B-Chat-v1.0) on the data/ui_math and the data/ui_coding datasets.
+It achieves the following results on the evaluation set:
+- Loss: 0.3968
+- Rewards/chosen: -1.5087
+- Rewards/rejected: -3.7229
+- Rewards/accuracies: 0.7812
+- Rewards/margins: 2.2142
+- Logps/rejected: -740.7106
+- Logps/chosen: -538.8605
+- Logits/rejected: -2.5209
+- Logits/chosen: -2.5392
 
 ## Model description
 

all_results.json

@@ -1,5 +1,18 @@
 {
     "epoch": 0.9963369963369964,
+    "eval_logits/chosen": -2.5392236709594727,
+    "eval_logits/rejected": -2.520923614501953,
+    "eval_logps/chosen": -538.8605346679688,
+    "eval_logps/rejected": -740.7106323242188,
+    "eval_loss": 0.39680731296539307,
+    "eval_rewards/accuracies": 0.78125,
+    "eval_rewards/chosen": -1.508690357208252,
+    "eval_rewards/margins": 2.2141778469085693,
+    "eval_rewards/rejected": -3.7228684425354004,
+    "eval_runtime": 2.9476,
+    "eval_samples": 200,
+    "eval_samples_per_second": 67.851,
+    "eval_steps_per_second": 0.679,
     "total_flos": 0.0,
     "train_loss": 0.4527332771058176,
     "train_runtime": 5381.1516,

config.json

@@ -24,6 +24,6 @@
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.44.1",
-  "use_cache": false,
+  "use_cache": true,
   "vocab_size": 32000
 }

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 0.9963369963369964,
+    "eval_logits/chosen": -2.5392236709594727,
+    "eval_logits/rejected": -2.520923614501953,
+    "eval_logps/chosen": -538.8605346679688,
+    "eval_logps/rejected": -740.7106323242188,
+    "eval_loss": 0.39680731296539307,
+    "eval_rewards/accuracies": 0.78125,
+    "eval_rewards/chosen": -1.508690357208252,
+    "eval_rewards/margins": 2.2141778469085693,
+    "eval_rewards/rejected": -3.7228684425354004,
+    "eval_runtime": 2.9476,
+    "eval_samples": 200,
+    "eval_samples_per_second": 67.851,
+    "eval_steps_per_second": 0.679
+}