End of training

README.md

@@ -1,4 +1,6 @@
 ---
+language:
+- en
 license: apache-2.0
 tags:
 - generated_from_trainer
@@ -13,7 +15,7 @@ model-index:
       name: Text Classification
       type: text-classification
     dataset:
-      name: glue
+      name: GLUE WNLI
       type: glue
       config: wnli
       split: validation
@@ -29,9 +31,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # distilbert_sa_GLUE_Experiment_logit_kd_wnli_96
 
-This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the glue dataset.
+This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the GLUE WNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.3453
+- Loss: 0.3441
 - Accuracy: 0.5634
 
 ## Model description

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 10.0,
+    "eval_accuracy": 0.5633802816901409,
+    "eval_loss": 0.3441176414489746,
+    "eval_runtime": 0.1266,
+    "eval_samples": 71,
+    "eval_samples_per_second": 560.81,
+    "eval_steps_per_second": 7.899,
+    "train_loss": 0.34743192195892336,
+    "train_runtime": 34.4199,
+    "train_samples": 635,
+    "train_samples_per_second": 922.431,
+    "train_steps_per_second": 4.358
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 10.0,
+    "eval_accuracy": 0.5633802816901409,
+    "eval_loss": 0.3441176414489746,
+    "eval_runtime": 0.1266,
+    "eval_samples": 71,
+    "eval_samples_per_second": 560.81,
+    "eval_steps_per_second": 7.899
+}

logs/events.out.tfevents.1675047684.serv-3328.2086075.23

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b95f0ff8349ffc0ab6c55ab6dd179ba9ee853f3db15c416778976841293732a3
+size 357

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 10.0,
+    "train_loss": 0.34743192195892336,
+    "train_runtime": 34.4199,
+    "train_samples": 635,
+    "train_samples_per_second": 922.431,
+    "train_steps_per_second": 4.358
+}

trainer_state.json

@@ -0,0 +1,175 @@
+{
+  "best_metric": 0.3441176414489746,
+  "best_model_checkpoint": "distilbert_sa_GLUE_Experiment_logit_kd_wnli_96/checkpoint-15",
+  "epoch": 10.0,
+  "global_step": 30,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.9e-05,
+      "loss": 0.348,
+      "step": 3
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.345082551240921,
+      "eval_runtime": 0.113,
+      "eval_samples_per_second": 628.265,
+      "eval_steps_per_second": 8.849,
+      "step": 3
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.8e-05,
+      "loss": 0.3477,
+      "step": 6
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.3447287976741791,
+      "eval_runtime": 0.1177,
+      "eval_samples_per_second": 603.052,
+      "eval_steps_per_second": 8.494,
+      "step": 6
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.7e-05,
+      "loss": 0.3467,
+      "step": 9
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.3444860577583313,
+      "eval_runtime": 0.1121,
+      "eval_samples_per_second": 633.302,
+      "eval_steps_per_second": 8.92,
+      "step": 9
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.3473,
+      "step": 12
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.34416645765304565,
+      "eval_runtime": 0.1108,
+      "eval_samples_per_second": 641.052,
+      "eval_steps_per_second": 9.029,
+      "step": 12
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 4.5e-05,
+      "loss": 0.3474,
+      "step": 15
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.3441176414489746,
+      "eval_runtime": 0.1102,
+      "eval_samples_per_second": 644.292,
+      "eval_steps_per_second": 9.075,
+      "step": 15
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.3476,
+      "step": 18
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.34430181980133057,
+      "eval_runtime": 0.1109,
+      "eval_samples_per_second": 640.012,
+      "eval_steps_per_second": 9.014,
+      "step": 18
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 4.3e-05,
+      "loss": 0.3477,
+      "step": 21
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.3445564806461334,
+      "eval_runtime": 0.1111,
+      "eval_samples_per_second": 638.78,
+      "eval_steps_per_second": 8.997,
+      "step": 21
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 4.2e-05,
+      "loss": 0.347,
+      "step": 24
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.3449000120162964,
+      "eval_runtime": 0.111,
+      "eval_samples_per_second": 639.483,
+      "eval_steps_per_second": 9.007,
+      "step": 24
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 4.1e-05,
+      "loss": 0.3477,
+      "step": 27
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.34510910511016846,
+      "eval_runtime": 0.1348,
+      "eval_samples_per_second": 526.598,
+      "eval_steps_per_second": 7.417,
+      "step": 27
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 4e-05,
+      "loss": 0.3472,
+      "step": 30
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.3452550768852234,
+      "eval_runtime": 0.1099,
+      "eval_samples_per_second": 645.899,
+      "eval_steps_per_second": 9.097,
+      "step": 30
+    },
+    {
+      "epoch": 10.0,
+      "step": 30,
+      "total_flos": 36999717191680.0,
+      "train_loss": 0.34743192195892336,
+      "train_runtime": 34.4199,
+      "train_samples_per_second": 922.431,
+      "train_steps_per_second": 4.358
+    }
+  ],
+  "max_steps": 150,
+  "num_train_epochs": 50,
+  "total_flos": 36999717191680.0,
+  "trial_name": null,
+  "trial_params": null
+}