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 QNLI
       type: glue
       config: qnli
       split: validation
@@ -21,7 +23,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.6174263225334066
+      value: 0.5883214351089145
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -29,10 +31,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # distilbert_add_GLUE_Experiment_logit_kd_qnli
 
-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 QNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.4068
-- Accuracy: 0.6174
+- Loss: 0.3978
+- Accuracy: 0.5883
 
 ## Model description
 

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 7.0,
+    "eval_accuracy": 0.5883214351089145,
+    "eval_loss": 0.39777371287345886,
+    "eval_runtime": 7.693,
+    "eval_samples": 5463,
+    "eval_samples_per_second": 710.13,
+    "eval_steps_per_second": 2.86,
+    "train_loss": 0.3814040659196701,
+    "train_runtime": 1634.9313,
+    "train_samples": 104743,
+    "train_samples_per_second": 3203.284,
+    "train_steps_per_second": 12.539
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 7.0,
+    "eval_accuracy": 0.5883214351089145,
+    "eval_loss": 0.39777371287345886,
+    "eval_runtime": 7.693,
+    "eval_samples": 5463,
+    "eval_samples_per_second": 710.13,
+    "eval_steps_per_second": 2.86
+}

logs/events.out.tfevents.1674948021.serv-3328.2403931.8

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7f9d8aff4f374538e9afcd110681e636fc9d11b0866c59dffadb98cd3faeaf2c
+size 363

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 7.0,
+    "train_loss": 0.3814040659196701,
+    "train_runtime": 1634.9313,
+    "train_samples": 104743,
+    "train_samples_per_second": 3203.284,
+    "train_steps_per_second": 12.539
+}

trainer_state.json

@@ -0,0 +1,130 @@
+{
+  "best_metric": 0.39777371287345886,
+  "best_model_checkpoint": "distilbert_add_GLUE_Experiment_logit_kd_qnli/checkpoint-820",
+  "epoch": 7.0,
+  "global_step": 2870,
+  "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": 0.4154,
+      "step": 410
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.5778876075416438,
+      "eval_loss": 0.39863064885139465,
+      "eval_runtime": 7.79,
+      "eval_samples_per_second": 701.281,
+      "eval_steps_per_second": 2.824,
+      "step": 410
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.8e-05,
+      "loss": 0.3986,
+      "step": 820
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.5883214351089145,
+      "eval_loss": 0.39777371287345886,
+      "eval_runtime": 7.6923,
+      "eval_samples_per_second": 710.188,
+      "eval_steps_per_second": 2.86,
+      "step": 820
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.7e-05,
+      "loss": 0.3909,
+      "step": 1230
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.5886875343218012,
+      "eval_loss": 0.39901256561279297,
+      "eval_runtime": 7.6905,
+      "eval_samples_per_second": 710.361,
+      "eval_steps_per_second": 2.861,
+      "step": 1230
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.384,
+      "step": 1640
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.5912502288120081,
+      "eval_loss": 0.39876481890678406,
+      "eval_runtime": 7.7111,
+      "eval_samples_per_second": 708.461,
+      "eval_steps_per_second": 2.853,
+      "step": 1640
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 4.5e-05,
+      "loss": 0.3761,
+      "step": 2050
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.5899688815669046,
+      "eval_loss": 0.400068461894989,
+      "eval_runtime": 7.7122,
+      "eval_samples_per_second": 708.359,
+      "eval_steps_per_second": 2.853,
+      "step": 2050
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.3634,
+      "step": 2460
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.6121178839465495,
+      "eval_loss": 0.40260177850723267,
+      "eval_runtime": 7.7119,
+      "eval_samples_per_second": 708.388,
+      "eval_steps_per_second": 2.853,
+      "step": 2460
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 4.3e-05,
+      "loss": 0.3413,
+      "step": 2870
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.6174263225334066,
+      "eval_loss": 0.40683838725090027,
+      "eval_runtime": 7.7369,
+      "eval_samples_per_second": 706.101,
+      "eval_steps_per_second": 2.844,
+      "step": 2870
+    },
+    {
+      "epoch": 7.0,
+      "step": 2870,
+      "total_flos": 4.469658863062221e+16,
+      "train_loss": 0.3814040659196701,
+      "train_runtime": 1634.9313,
+      "train_samples_per_second": 3203.284,
+      "train_steps_per_second": 12.539
+    }
+  ],
+  "max_steps": 20500,
+  "num_train_epochs": 50,
+  "total_flos": 4.469658863062221e+16,
+  "trial_name": null,
+  "trial_params": null
+}