Model save

README.md

@@ -1,11 +1,9 @@
 ---
 base_model: openai-community/gpt2
-datasets: HuggingFaceH4/Bespoke-Stratos-17k
 library_name: transformers
 model_name: gpt2-SFT
 tags:
 - generated_from_trainer
-- open-r1
 - trl
 - sft
 licence: license
@@ -13,7 +11,7 @@ licence: license
 
 # Model Card for gpt2-SFT
 
-This model is a fine-tuned version of [openai-community/gpt2](https://huggingface.co/openai-community/gpt2) on the [HuggingFaceH4/Bespoke-Stratos-17k](https://huggingface.co/datasets/HuggingFaceH4/Bespoke-Stratos-17k) dataset.
+This model is a fine-tuned version of [openai-community/gpt2](https://huggingface.co/openai-community/gpt2).
 It has been trained using [TRL](https://github.com/huggingface/trl).
 
 ## Quick start
@@ -29,14 +27,14 @@ 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/s18574s18574-/huggingface/runs/eptavvvk) 
+[<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/s18574s18574-/huggingface/runs/92j1a6yg) 
 
 
 This model was trained with SFT.
 
 ### Framework versions
 
-- TRL: 0.16.0.dev0
+- TRL: 0.15.2
 - Transformers: 4.50.0.dev0
 - Pytorch: 2.5.1
 - Datasets: 3.3.2

all_results.json

@@ -4,10 +4,10 @@
     "eval_samples": 100,
     "eval_samples_per_second": 239.436,
     "eval_steps_per_second": 60.186,
-    "total_flos": 4.794029428021658e+16,
-    "train_loss": 1.842788679657142,
-    "train_runtime": 1095.7674,
-    "train_samples": 16610,
-    "train_samples_per_second": 83.719,
-    "train_steps_per_second": 2.616
+    "total_flos": 15459466346496.0,
+    "train_loss": 3.207212707142771,
+    "train_runtime": 884.063,
+    "train_samples": 19883,
+    "train_samples_per_second": 23.43,
+    "train_steps_per_second": 0.366
 }

config.json

@@ -1,5 +1,4 @@
 {
-  "_name_or_path": "openai-community/gpt2",
   "activation_function": "gelu_new",
   "architectures": [
     "GPT2LMHeadModel"
@@ -34,6 +33,6 @@
   },
   "torch_dtype": "bfloat16",
   "transformers_version": "4.50.0.dev0",
-  "use_cache": true,
+  "use_cache": false,
   "vocab_size": 50257
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d7a4d3e9366758ab67ab6631e5f3cf745e2ecf43c3dcbaa713d59efbb4912848
-size 326089656
+oid sha256:1f42938a2cbb5158523dd7829a5530450f1a8f4d17462bdb98f4adbfa962c0cc
+size 248894656

tokenizer_config.json

@@ -11,7 +11,6 @@
     }
   },
   "bos_token": "<|endoftext|>",
-  "chat_template": "{% for message in messages %}{{ message['role'] }}: {{ message['content'] }}\n{% endfor %}",
   "clean_up_tokenization_spaces": false,
   "eos_token": "<|endoftext|>",
   "extra_special_tokens": {},

train_results.json

@@ -1,8 +1,8 @@
 {
-    "total_flos": 4.794029428021658e+16,
-    "train_loss": 1.842788679657142,
-    "train_runtime": 1095.7674,
-    "train_samples": 16610,
-    "train_samples_per_second": 83.719,
-    "train_steps_per_second": 2.616
+    "total_flos": 15459466346496.0,
+    "train_loss": 3.207212707142771,
+    "train_runtime": 884.063,
+    "train_samples": 19883,
+    "train_samples_per_second": 23.43,
+    "train_steps_per_second": 0.366
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:895d105dd61e8ebfe31e8e0469aec7ef17c2b3df2ae35e198d4214d31665f4d7
-size 7352
+oid sha256:047c9541a64228bde09a78f0ca1cf98bb5e12b59574cf6d5c1d47e96a8bc50e1
+size 7288