End of training

README.md

@@ -4,6 +4,7 @@ license: apache-2.0
 base_model: mistralai/Mistral-7B-Instruct-v0.3
 tags:
 - llama-factory
+- ia3
 - generated_from_trainer
 datasets:
 - super_glue
@@ -17,9 +18,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # train_boolq_1745950279
 
-This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.3](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.3) on the super_glue dataset.
+This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.3](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.3) on the boolq dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.1744
+- Loss: 0.1734
 - Num Input Tokens Seen: 37097424
 
 ## Model description

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 18.85902876001886,
+    "eval_loss": 0.17341691255569458,
+    "eval_runtime": 31.4035,
+    "eval_samples_per_second": 30.028,
+    "eval_steps_per_second": 15.03,
+    "num_input_tokens_seen": 37097424,
+    "total_flos": 1.583466987341611e+18,
+    "train_loss": 0.1640541000654921,
+    "train_runtime": 26823.2699,
+    "train_samples_per_second": 5.965,
+    "train_steps_per_second": 1.491
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 18.85902876001886,
+    "eval_loss": 0.17341691255569458,
+    "eval_runtime": 31.4035,
+    "eval_samples_per_second": 30.028,
+    "eval_steps_per_second": 15.03,
+    "num_input_tokens_seen": 37097424
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 18.85902876001886,
+    "num_input_tokens_seen": 37097424,
+    "total_flos": 1.583466987341611e+18,
+    "train_loss": 0.1640541000654921,
+    "train_runtime": 26823.2699,
+    "train_samples_per_second": 5.965,
+    "train_steps_per_second": 1.491
+}