Model save

README.md

@@ -4,7 +4,6 @@ library_name: transformers
 model_name: zephyr-7b-dpo-full
 tags:
 - generated_from_trainer
-- alignment-handbook
 - trl
 - dpo
 licence: license
@@ -28,7 +27,7 @@ print(output["generated_text"])
 
 ## Training procedure
 
-[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/devamanyu/huggingface/runs/adglv44w)
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/devamanyu/huggingface/runs/2spwkmbc)
 
 This model was trained with DPO, a method introduced in [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://huggingface.co/papers/2305.18290).
 

all_results.json

@@ -15,8 +15,8 @@
     "eval_steps_per_second": 0.8,
     "total_flos": 0.0,
     "train_loss": 0.0,
-    "train_runtime": 1.2352,
+    "train_runtime": 1.1955,
     "train_samples": 100,
-    "train_samples_per_second": 8095.94,
-    "train_steps_per_second": 80.959
+    "train_samples_per_second": 8364.905,
+    "train_steps_per_second": 83.649
 }

config.json

@@ -22,6 +22,6 @@
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.46.1",
-  "use_cache": true,
+  "use_cache": false,
   "vocab_size": 32000
 }

runs/Nov22_20-16-58_dvhaz-sleeper-cmh-gpu/events.out.tfevents.1732306662.dvhaz-sleeper-cmh-gpu.10458.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7c0fa702483a32cd212fb4eff6fb1ae60473c6b9a617c9c5703661321108a17
+size 6760

train_results.json

@@ -2,8 +2,8 @@
     "epoch": 1.0,
     "total_flos": 0.0,
     "train_loss": 0.0,
-    "train_runtime": 1.2352,
+    "train_runtime": 1.1955,
     "train_samples": 100,
-    "train_samples_per_second": 8095.94,
-    "train_steps_per_second": 80.959
+    "train_samples_per_second": 8364.905,
+    "train_steps_per_second": 83.649
 }

trainer_state.json

@@ -797,9 +797,9 @@
       "step": 478,
       "total_flos": 0.0,
       "train_loss": 0.0,
-      "train_runtime": 1.2352,
-      "train_samples_per_second": 8095.94,
-      "train_steps_per_second": 80.959
+      "train_runtime": 1.1955,
+      "train_samples_per_second": 8364.905,
+      "train_steps_per_second": 83.649
     }
   ],
   "logging_steps": 10,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9a8158ba2af669bd22aca4ac8a83903e60cd43f83e0a28dd5cb7139946725296
+oid sha256:5493cf08969e29ad06ed8c2bb6e5927f0f327e31cccd994af7be298bb5852953
 size 7736