End of training

README.md

@@ -4,6 +4,7 @@ license: apache-2.0
 base_model: mistralai/Mistral-7B-Instruct-v0.3
 tags:
 - llama-factory
+- prompt-tuning
 - generated_from_trainer
 model-index:
 - name: train_rte_1744902663
@@ -15,9 +16,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # train_rte_1744902663
 
-This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.3](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.3) on an unknown 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 rte dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.7575
+- Loss: 0.1400
 - Num Input Tokens Seen: 107274480
 
 ## Model description

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 283.6916221033868,
+    "eval_loss": 0.14000995457172394,
+    "eval_runtime": 6.5223,
+    "eval_samples_per_second": 38.177,
+    "eval_steps_per_second": 9.659,
+    "num_input_tokens_seen": 107274480,
+    "total_flos": 4.578778923543429e+18,
+    "train_loss": 0.09070941930053177,
+    "train_runtime": 36100.9279,
+    "train_samples_per_second": 17.728,
+    "train_steps_per_second": 1.108
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 283.6916221033868,
+    "eval_loss": 0.14000995457172394,
+    "eval_runtime": 6.5223,
+    "eval_samples_per_second": 38.177,
+    "eval_steps_per_second": 9.659,
+    "num_input_tokens_seen": 107274480
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 283.6916221033868,
+    "num_input_tokens_seen": 107274480,
+    "total_flos": 4.578778923543429e+18,
+    "train_loss": 0.09070941930053177,
+    "train_runtime": 36100.9279,
+    "train_samples_per_second": 17.728,
+    "train_steps_per_second": 1.108
+}