End of training

README.md

@@ -3,11 +3,15 @@ library_name: transformers
 license: llama3
 base_model: meta-llama/Meta-Llama-3-8B
 tags:
+- alignment-handbook
+- trl
+- sft
+- generated_from_trainer
 - trl
 - sft
 - generated_from_trainer
 datasets:
-- generator
+- llama-duo/synth_summarize_dataset_dedup
 model-index:
 - name: llama3.1-8b-gpt4o_100k_summarize-fft
   results: []
@@ -18,7 +22,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # llama3.1-8b-gpt4o_100k_summarize-fft
 
-This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) on the generator dataset.
+This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) on the llama-duo/synth_summarize_dataset_dedup dataset.
 It achieves the following results on the evaluation set:
 - Loss: 3.8606
 

all_results.json

@@ -1,5 +1,10 @@
 {
     "epoch": 1.0,
+    "eval_loss": 3.8606326580047607,
+    "eval_runtime": 2.1076,
+    "eval_samples": 25,
+    "eval_samples_per_second": 4.745,
+    "eval_steps_per_second": 0.474,
     "total_flos": 20388343971840.0,
     "train_loss": 1.4792061909651144,
     "train_runtime": 8464.1713,

config.json

@@ -26,6 +26,6 @@
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.45.1",
-  "use_cache": false,
+  "use_cache": true,
   "vocab_size": 128258
 }

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.0,
+    "eval_loss": 3.8606326580047607,
+    "eval_runtime": 2.1076,
+    "eval_samples": 25,
+    "eval_samples_per_second": 4.745,
+    "eval_steps_per_second": 0.474
+}

runs/Sep27_21-33-16_637888cb4368/events.out.tfevents.1727481947.637888cb4368.21341.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:657cb79a2489f52bf867c22fa2272799b263b6b602aa3d7b643926d304518cea
+size 359