End of training

README.md

@@ -18,9 +18,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # train_hellaswag_101112_1760638083
 
-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 an unknown 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 hellaswag dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.2429
+- Loss: 0.0606
 - Num Input Tokens Seen: 218373904
 
 ## Model description

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 20.0,
+    "eval_loss": 0.06061485782265663,
+    "eval_runtime": 124.1275,
+    "eval_samples_per_second": 32.152,
+    "eval_steps_per_second": 8.04,
+    "num_input_tokens_seen": 218373904,
+    "total_flos": 9.852512656423256e+18,
+    "train_loss": 0.015886625606237346,
+    "train_runtime": 78418.9489,
+    "train_samples_per_second": 9.16,
+    "train_steps_per_second": 2.29
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 20.0,
+    "eval_loss": 0.06061485782265663,
+    "eval_runtime": 124.1275,
+    "eval_samples_per_second": 32.152,
+    "eval_steps_per_second": 8.04,
+    "num_input_tokens_seen": 218373904
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 20.0,
+    "num_input_tokens_seen": 218373904,
+    "total_flos": 9.852512656423256e+18,
+    "train_loss": 0.015886625606237346,
+    "train_runtime": 78418.9489,
+    "train_samples_per_second": 9.16,
+    "train_steps_per_second": 2.29
+}