Upload folder using huggingface_hub

README.md

@@ -32,7 +32,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [facebook/opt-350m](https://huggingface.co/facebook/opt-350m) on the GLUE RTE dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.1042
+- Loss: 2.0566
 - Accuracy: 0.7112
 
 ## Model description
@@ -53,7 +53,7 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 2e-05
-- train_batch_size: 32
+- train_batch_size: 16
 - eval_batch_size: 8
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08

all_results.json

@@ -1,14 +1,14 @@
 {
     "epoch": 5.0,
     "eval_accuracy": 0.7111913357400722,
-    "eval_loss": 2.1042087078094482,
-    "eval_runtime": 1.5849,
+    "eval_loss": 2.056576728820801,
+    "eval_runtime": 0.784,
     "eval_samples": 277,
-    "eval_samples_per_second": 174.772,
-    "eval_steps_per_second": 22.083,
-    "train_loss": 0.30048399705153245,
-    "train_runtime": 200.6393,
+    "eval_samples_per_second": 353.301,
+    "eval_steps_per_second": 44.641,
+    "train_loss": 0.321053017102755,
+    "train_runtime": 84.2492,
     "train_samples": 2490,
-    "train_samples_per_second": 62.052,
-    "train_steps_per_second": 1.944
+    "train_samples_per_second": 147.776,
+    "train_steps_per_second": 9.258
 }

eval_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 5.0,
     "eval_accuracy": 0.7111913357400722,
-    "eval_loss": 2.1042087078094482,
-    "eval_runtime": 1.5849,
+    "eval_loss": 2.056576728820801,
+    "eval_runtime": 0.784,
     "eval_samples": 277,
-    "eval_samples_per_second": 174.772,
-    "eval_steps_per_second": 22.083
+    "eval_samples_per_second": 353.301,
+    "eval_steps_per_second": 44.641
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b358536980e2f2c0deac46a206a0542b52e165e42423a83c7316fe8553ef30c7
+oid sha256:70978ed91bc8237a364f1abc9663e9a21c1845bf6e97bd2a8648f64ca83c3d1e
 size 1324921569

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 5.0,
-    "train_loss": 0.30048399705153245,
-    "train_runtime": 200.6393,
+    "train_loss": 0.321053017102755,
+    "train_runtime": 84.2492,
     "train_samples": 2490,
-    "train_samples_per_second": 62.052,
-    "train_steps_per_second": 1.944
+    "train_samples_per_second": 147.776,
+    "train_steps_per_second": 9.258
 }

trainer_state.json

@@ -2,22 +2,28 @@
   "best_metric": null,
   "best_model_checkpoint": null,
   "epoch": 5.0,
-  "global_step": 390,
+  "global_step": 780,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
+    {
+      "epoch": 3.21,
+      "learning_rate": 7.282051282051282e-06,
+      "loss": 0.47,
+      "step": 500
+    },
     {
       "epoch": 5.0,
-      "step": 390,
+      "step": 780,
       "total_flos": 2900597184921600.0,
-      "train_loss": 0.30048399705153245,
-      "train_runtime": 200.6393,
-      "train_samples_per_second": 62.052,
-      "train_steps_per_second": 1.944
+      "train_loss": 0.321053017102755,
+      "train_runtime": 84.2492,
+      "train_samples_per_second": 147.776,
+      "train_steps_per_second": 9.258
     }
   ],
-  "max_steps": 390,
+  "max_steps": 780,
   "num_train_epochs": 5,
   "total_flos": 2900597184921600.0,
   "trial_name": null,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3b3b22874f2da3cca2923e809d12e9c08b60ee574321286e71bb5196c3407c8e
+oid sha256:59537fe8b1e8f4d9b7dbdb8d5995dcdb60103dcce782bee474e7b5ed99736261
 size 3963