End of training

README.md

@@ -17,7 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # train_winogrande_101112_1760638068
 
-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 winogrande dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.2313
 - Num Input Tokens Seen: 38366624

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 20.0,
+    "eval_loss": 0.23126927018165588,
+    "eval_runtime": 60.3272,
+    "eval_samples_per_second": 66.968,
+    "eval_steps_per_second": 16.742,
+    "num_input_tokens_seen": 38366624,
+    "total_flos": 1.727631738233684e+18,
+    "train_loss": 0.23324829005598366,
+    "train_runtime": 25757.3665,
+    "train_samples_per_second": 28.231,
+    "train_steps_per_second": 7.058
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 20.0,
+    "eval_loss": 0.23126927018165588,
+    "eval_runtime": 60.3272,
+    "eval_samples_per_second": 66.968,
+    "eval_steps_per_second": 16.742,
+    "num_input_tokens_seen": 38366624
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 20.0,
+    "num_input_tokens_seen": 38366624,
+    "total_flos": 1.727631738233684e+18,
+    "train_loss": 0.23324829005598366,
+    "train_runtime": 25757.3665,
+    "train_samples_per_second": 28.231,
+    "train_steps_per_second": 7.058
+}