End of training

README.md

@@ -1,9 +1,16 @@
 ---
 library_name: transformers
+base_model: data/OpenELM-1_1B-SFT-1
 tags:
+- alignment-handbook
 - trl
 - dpo
 - generated_from_trainer
+- trl
+- dpo
+- generated_from_trainer
+datasets:
+- HuggingFaceH4/ultrafeedback_binarized
 model-index:
 - name: OpenELM-1_1B-DPO-full-1
   results: []
@@ -14,12 +21,12 @@ should probably proofread and complete it, then remove this comment. -->
 
 # OpenELM-1_1B-DPO-full-1
 
-This model was trained from scratch on an unknown dataset.
+This model is a fine-tuned version of [data/OpenELM-1_1B-SFT-1](https://huggingface.co/data/OpenELM-1_1B-SFT-1) on the HuggingFaceH4/ultrafeedback_binarized dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.8121
+- Loss: 0.8127
 - Rewards/chosen: -7.4062
 - Rewards/rejected: -9.625
-- Rewards/accuracies: 0.7285
+- Rewards/accuracies: 0.7266
 - Rewards/margins: 2.2188
 - Logps/rejected: -1248.0
 - Logps/chosen: -1056.0

all_results.json

@@ -1,5 +1,18 @@
 {
     "epoch": 2.998430141287284,
+    "eval_logits/chosen": -4.0,
+    "eval_logits/rejected": -1.578125,
+    "eval_logps/chosen": -1056.0,
+    "eval_logps/rejected": -1248.0,
+    "eval_loss": 0.812749981880188,
+    "eval_rewards/accuracies": 0.7265625,
+    "eval_rewards/chosen": -7.40625,
+    "eval_rewards/margins": 2.21875,
+    "eval_rewards/rejected": -9.625,
+    "eval_runtime": 46.6963,
+    "eval_samples": 2000,
+    "eval_samples_per_second": 42.83,
+    "eval_steps_per_second": 0.685,
     "total_flos": 0.0,
     "train_loss": 0.26550976372930185,
     "train_runtime": 11897.7673,

config.json

@@ -120,6 +120,6 @@
   "share_input_output_layers": true,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.44.2",
-  "use_cache": false,
+  "use_cache": true,
   "vocab_size": 32000
 }

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 2.998430141287284,
+    "eval_logits/chosen": -4.0,
+    "eval_logits/rejected": -1.578125,
+    "eval_logps/chosen": -1056.0,
+    "eval_logps/rejected": -1248.0,
+    "eval_loss": 0.812749981880188,
+    "eval_rewards/accuracies": 0.7265625,
+    "eval_rewards/chosen": -7.40625,
+    "eval_rewards/margins": 2.21875,
+    "eval_rewards/rejected": -9.625,
+    "eval_runtime": 46.6963,
+    "eval_samples": 2000,
+    "eval_samples_per_second": 42.83,
+    "eval_steps_per_second": 0.685
+}

runs/Aug29_16-13-04_xe8545-a100-12/events.out.tfevents.1724953229.xe8545-a100-12.526559.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:74c964e9e101ea5372e17f4d8d4c32f1c8569752fedc0d50ed66c578c77eddef
+size 828