End of training

README.md

@@ -1,11 +1,15 @@
 ---
 base_model: meta-llama/Llama-2-7b-hf
 tags:
+- alignment-handbook
+- trl
+- sft
+- generated_from_trainer
 - trl
 - sft
 - generated_from_trainer
 datasets:
-- generator
+- Lichang-Chen/800k_ift
 model-index:
 - name: llama2-ift-800k
   results: []
@@ -16,7 +20,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # llama2-ift-800k
 
-This model is a fine-tuned version of [meta-llama/Llama-2-7b-hf](https://huggingface.co/meta-llama/Llama-2-7b-hf) on the generator dataset.
+This model is a fine-tuned version of [meta-llama/Llama-2-7b-hf](https://huggingface.co/meta-llama/Llama-2-7b-hf) on the Lichang-Chen/800k_ift dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.6256
 

all_results.json

@@ -1,5 +1,10 @@
 {
     "epoch": 3.0,
+    "eval_loss": 0.6255540251731873,
+    "eval_runtime": 171.1498,
+    "eval_samples": 41854,
+    "eval_samples_per_second": 48.706,
+    "eval_steps_per_second": 0.765,
     "train_loss": 0.6186258538300636,
     "train_runtime": 37714.0323,
     "train_samples": 795212,

config.json

@@ -23,6 +23,6 @@
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.39.0.dev0",
-  "use_cache": false,
+  "use_cache": true,
   "vocab_size": 32000
 }

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "eval_loss": 0.6255540251731873,
+    "eval_runtime": 171.1498,
+    "eval_samples": 41854,
+    "eval_samples_per_second": 48.706,
+    "eval_steps_per_second": 0.765
+}

runs/Apr15_16-17-45_a100-80-8-zifeng-2/events.out.tfevents.1713238466.a100-80-8-zifeng-2.3327053.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a8bc900a9b2098c7d6d65ab8edb70bb0648621bc93700c543dc374d159fd4cc1
+size 359