Training in progress, epoch 3

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:806bf53cb67d164eceb27c2bda7203b50c1bc8336ad79beead3992c948e3f3e0
+oid sha256:1c8091a6d50c16ffaec776032560df90efcaf84d3ef650fe1893203fe2c3b07e
 size 267832560

wandb/run-20240924_172630-x9iddikd/files/output.log

@@ -21,3 +21,4 @@ Some weights of DistilBertForSequenceClassification were not initialized from th
 You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.
 {'eval_loss': 0.21834564208984375, 'eval_accuracy': 0.938, 'eval_runtime': 30.3683, 'eval_samples_per_second': 32.929, 'eval_steps_per_second': 2.075, 'epoch': 1.0}
 {'loss': 0.2031, 'grad_norm': 1.1480563879013062, 'learning_rate': 1.2e-05, 'epoch': 2.0}
+{'eval_loss': 0.19427122175693512, 'eval_accuracy': 0.938, 'eval_runtime': 42.2287, 'eval_samples_per_second': 23.681, 'eval_steps_per_second': 1.492, 'epoch': 2.0}

wandb/run-20240924_172630-x9iddikd/files/wandb-summary.json

@@ -1 +1 @@
-{"eval/loss": 0.19427122175693512, "eval/accuracy": 0.938, "eval/runtime": 42.2287, "eval/samples_per_second": 23.681, "eval/steps_per_second": 1.492, "train/epoch": 2.0, "train/global_step": 500, "_timestamp": 1727212675.9347782, "_runtime": 3085.0618691444397, "_step": 2, "train/loss": 0.2031, "train/grad_norm": 1.1480563879013062, "train/learning_rate": 1.2e-05}
+{"eval/loss": 0.3195326626300812, "eval/accuracy": 0.921, "eval/runtime": 26.5577, "eval/samples_per_second": 37.654, "eval/steps_per_second": 2.372, "train/epoch": 3.0, "train/global_step": 750, "_timestamp": 1727214317.037596, "_runtime": 4726.164686918259, "_step": 3, "train/loss": 0.2031, "train/grad_norm": 1.1480563879013062, "train/learning_rate": 1.2e-05}