End of training

README.md

@@ -2,11 +2,15 @@
 license: llama3
 base_model: meta-llama/Meta-Llama-3-8B-Instruct
 tags:
+- alignment-handbook
+- trl
+- sft
+- generated_from_trainer
 - trl
 - sft
 - generated_from_trainer
 datasets:
-- generator
+- HuggingFaceH4/deita-10k-v0-sft
 model-index:
 - name: llama3-8b-instruct-wo-kqa_golden-iter-sft-step1
   results: []
@@ -17,7 +21,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # llama3-8b-instruct-wo-kqa_golden-iter-sft-step1
 
-This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on the generator dataset.
+This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on the HuggingFaceH4/deita-10k-v0-sft dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.4835
 

all_results.json

@@ -1,5 +1,10 @@
 {
     "epoch": 3.0,
+    "eval_loss": 0.48352566361427307,
+    "eval_runtime": 3.6883,
+    "eval_samples": 201,
+    "eval_samples_per_second": 7.863,
+    "eval_steps_per_second": 0.542,
     "train_loss": 0.44834805979873193,
     "train_runtime": 748.259,
     "train_samples": 4747,

config.json

@@ -23,6 +23,6 @@
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.38.2",
-  "use_cache": false,
+  "use_cache": true,
   "vocab_size": 128256
 }

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "eval_loss": 0.48352566361427307,
+    "eval_runtime": 3.6883,
+    "eval_samples": 201,
+    "eval_samples_per_second": 7.863,
+    "eval_steps_per_second": 0.542
+}