End of training

README.md

@@ -4,6 +4,7 @@ license: apache-2.0
 base_model: mistralai/Mistral-7B-Instruct-v0.3
 tags:
 - llama-factory
+- lntuning
 - generated_from_trainer
 datasets:
 - super_glue
@@ -17,9 +18,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # train_cb_1745950320
 
-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 cb dataset.
 It achieves the following results on the evaluation set:
-- Loss: 3.5374
+- Loss: 3.5088
 - Num Input Tokens Seen: 23078128
 
 ## Model description

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 701.7610619469026,
+    "eval_loss": 3.5087850093841553,
+    "eval_runtime": 1.1202,
+    "eval_samples_per_second": 22.318,
+    "eval_steps_per_second": 11.605,
+    "num_input_tokens_seen": 23078128,
+    "total_flos": 9.850767941652972e+17,
+    "train_loss": 3.1605935114860535,
+    "train_runtime": 11642.513,
+    "train_samples_per_second": 13.743,
+    "train_steps_per_second": 3.436
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 701.7610619469026,
+    "eval_loss": 3.5087850093841553,
+    "eval_runtime": 1.1202,
+    "eval_samples_per_second": 22.318,
+    "eval_steps_per_second": 11.605,
+    "num_input_tokens_seen": 23078128
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 701.7610619469026,
+    "num_input_tokens_seen": 23078128,
+    "total_flos": 9.850767941652972e+17,
+    "train_loss": 3.1605935114860535,
+    "train_runtime": 11642.513,
+    "train_samples_per_second": 13.743,
+    "train_steps_per_second": 3.436
+}