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_closed_qa_dataset_dedup
 model-index:
 - name: llama3.1-8b-gpt4o_100k_closedqa-fft
   results: []
@@ -18,7 +22,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # llama3.1-8b-gpt4o_100k_closedqa-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_closed_qa_dataset_dedup dataset.
 It achieves the following results on the evaluation set:
 - Loss: 3.5964
 

all_results.json

@@ -1,5 +1,10 @@
 {
     "epoch": 0.9990224828934506,
+    "eval_loss": 3.596351385116577,
+    "eval_runtime": 2.1019,
+    "eval_samples": 15,
+    "eval_samples_per_second": 2.855,
+    "eval_steps_per_second": 0.476,
     "total_flos": 26722078556160.0,
     "train_loss": 1.3314139091805235,
     "train_runtime": 8881.3456,

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": 0.9990224828934506,
+    "eval_loss": 3.596351385116577,
+    "eval_runtime": 2.1019,
+    "eval_samples": 15,
+    "eval_samples_per_second": 2.855,
+    "eval_steps_per_second": 0.476
+}

runs/Sep28_01-51-51_637888cb4368/events.out.tfevents.1727498543.637888cb4368.138934.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0bba79d793196b05b9d49339d9f0d9fdf51d8b936a89e8ee12d34329f0e7baf7
+size 359