End of training

README.md

@@ -2,11 +2,15 @@
 license: mit
 base_model: openai-community/gpt2-medium
 tags:
+- alignment-handbook
+- trl
+- sft
+- generated_from_trainer
 - trl
 - sft
 - generated_from_trainer
 datasets:
-- generator
+- Anthropic/hh-rlhf
 model-index:
 - name: gpt2-sft
   results: []
@@ -17,7 +21,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # gpt2-sft
 
-This model is a fine-tuned version of [openai-community/gpt2-medium](https://huggingface.co/openai-community/gpt2-medium) on the generator dataset.
+This model is a fine-tuned version of [openai-community/gpt2-medium](https://huggingface.co/openai-community/gpt2-medium) on the Anthropic/hh-rlhf dataset.
 It achieves the following results on the evaluation set:
 - Loss: 2.1977
 

all_results.json

@@ -1,5 +1,10 @@
 {
     "epoch": 1.0,
+    "eval_loss": 2.1976709365844727,
+    "eval_runtime": 129.9307,
+    "eval_samples": 16080,
+    "eval_samples_per_second": 21.358,
+    "eval_steps_per_second": 2.671,
     "train_loss": 0.6064129045034048,
     "train_runtime": 690.5421,
     "train_samples": 64320,

config.json

@@ -36,6 +36,6 @@
   },
   "torch_dtype": "bfloat16",
   "transformers_version": "4.39.0.dev0",
-  "use_cache": false,
+  "use_cache": true,
   "vocab_size": 50257
 }

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.0,
+    "eval_loss": 2.1976709365844727,
+    "eval_runtime": 129.9307,
+    "eval_samples": 16080,
+    "eval_samples_per_second": 21.358,
+    "eval_steps_per_second": 2.671
+}

runs/Apr10_04-50-43_compsci-cluster-fitz-05/events.out.tfevents.1712739895.compsci-cluster-fitz-05.898746.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:73e569f3154116e5d7f2e1b8a9b693057190052bc87d0a77e5b8b3003e7e4a5f
+size 359