Auto upload zain 2026-08-18T15:42:29.977679 (part 3)

zain/Activation/wandb/run-20260818_154123-mv5l58cq/files/output.log

@@ -22,8 +22,43 @@ Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 50.
   - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
   - If you are not the owner of the model architecture class, please contact the model code owner to update it.
 Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 146.87it/s]
- 11%|█▏        | 286/2500 [00:29<01:41, 21.83it/s]?, ?it/s]
+ 12%|█▏        | 300/2500 [00:38<01:43, 21.18it[transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
 {'loss': '6.431', 'grad_norm': '1.703', 'learning_rate': '0.0001314', 'epoch': '0.007415', 'train/total_time_seconds': '5.195', 'train/time_per_step_avg': '0.02123', 'train/epoch_time_elapsed': '26.88', 'train/estimated_remaining_minutes': '0.8973'}
 {'loss': '6.208', 'grad_norm': '1.18', 'learning_rate': '0.0001434', 'epoch': '0.008089', 'train/total_time_seconds': '5.61', 'train/time_per_step_avg': '0.02121', 'train/epoch_time_elapsed': '27.79', 'train/estimated_remaining_minutes': '0.8805'}
 {'loss': '5.999', 'grad_norm': '1.125', 'learning_rate': '0.0001554', 'epoch': '0.008763', 'train/total_time_seconds': '6.03', 'train/time_per_step_avg': '0.02121', 'train/epoch_time_elapsed': '28.7', 'train/estimated_remaining_minutes': '0.8659'}
 {'loss': '5.795', 'grad_norm': '1.062', 'learning_rate': '0.0001674', 'epoch': '0.009437', 'train/total_time_seconds': '6.449', 'train/time_per_step_avg': '0.02116', 'train/epoch_time_elapsed': '29.61', 'train/estimated_remaining_minutes': '0.8522'}
+{'loss': '5.597', 'grad_norm': '0.9336', 'learning_rate': '0.0001794', 'epoch': '0.01011', 'train/total_time_seconds': '6.879', 'train/time_per_step_avg': '0.021', 'train/epoch_time_elapsed': '30.56', 'train/estimated_remaining_minutes': '0.8407'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+{'eval_loss': '5.48', 'eval_runtime': '7.865', 'eval_samples_per_second': '1211', 'eval_steps_per_second': '1.271', 'epoch': '0.01011', 'train/total_time_seconds': '6.879', 'train/time_per_step_avg': '0.021', 'train/epoch_time_elapsed': '38.43', 'train/estimated_remaining_minutes': '0.8407'}
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 160.47it/s]
+ 16%|█▌        | 400/2500 [00:51<01:41, 20.73it[transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '5.385', 'grad_norm': '0.9414', 'learning_rate': '0.0001914', 'epoch': '0.01079', 'train/total_time_seconds': '7.318', 'train/time_per_step_avg': '0.02124', 'train/epoch_time_elapsed': '39.41', 'train/estimated_remaining_minutes': '0.8309'}
+{'loss': '5.217', 'grad_norm': '0.9883', 'learning_rate': '0.0002034', 'epoch': '0.01146', 'train/total_time_seconds': '7.745', 'train/time_per_step_avg': '0.02134', 'train/epoch_time_elapsed': '40.35', 'train/estimated_remaining_minutes': '0.82'}
+{'loss': '5.037', 'grad_norm': '1.078', 'learning_rate': '0.0002154', 'epoch': '0.01213', 'train/total_time_seconds': '8.162', 'train/time_per_step_avg': '0.02132', 'train/epoch_time_elapsed': '41.26', 'train/estimated_remaining_minutes': '0.8087'}
+{'loss': '4.902', 'grad_norm': '1.211', 'learning_rate': '0.0002274', 'epoch': '0.01281', 'train/total_time_seconds': '8.576', 'train/time_per_step_avg': '0.02127', 'train/epoch_time_elapsed': '42.17', 'train/estimated_remaining_minutes': '0.7974'}
+{'loss': '4.778', 'grad_norm': '0.9844', 'learning_rate': '0.0002394', 'epoch': '0.01348', 'train/total_time_seconds': '9.008', 'train/time_per_step_avg': '0.0213', 'train/epoch_time_elapsed': '43.12', 'train/estimated_remaining_minutes': '0.7882'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+{'eval_loss': '4.718', 'eval_runtime': '7.988', 'eval_samples_per_second': '1193', 'eval_steps_per_second': '1.252', 'epoch': '0.01348', 'train/total_time_seconds': '9.008', 'train/time_per_step_avg': '0.0213', 'train/epoch_time_elapsed': '51.11', 'train/estimated_remaining_minutes': '0.7882'}
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 158.26it/s]
+ 20%|██        | 500/2500 [01:03<01:31, 21.96it[transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '4.657', 'grad_norm': '1.211', 'learning_rate': '0.0002514', 'epoch': '0.01416', 'train/total_time_seconds': '9.425', 'train/time_per_step_avg': '0.02107', 'train/epoch_time_elapsed': '52.06', 'train/estimated_remaining_minutes': '0.778'}
+{'loss': '4.587', 'grad_norm': '1.719', 'learning_rate': '0.0002634', 'epoch': '0.01483', 'train/total_time_seconds': '9.845', 'train/time_per_step_avg': '0.021', 'train/epoch_time_elapsed': '52.97', 'train/estimated_remaining_minutes': '0.7682'}
+{'loss': '4.508', 'grad_norm': '1.219', 'learning_rate': '0.0002754', 'epoch': '0.0155', 'train/total_time_seconds': '10.26', 'train/time_per_step_avg': '0.02102', 'train/epoch_time_elapsed': '53.89', 'train/estimated_remaining_minutes': '0.7587'}
+{'loss': '4.41', 'grad_norm': '1.25', 'learning_rate': '0.0002874', 'epoch': '0.01618', 'train/total_time_seconds': '10.7', 'train/time_per_step_avg': '0.0212', 'train/epoch_time_elapsed': '54.81', 'train/estimated_remaining_minutes': '0.7502'}
+{'loss': '4.358', 'grad_norm': '1.32', 'learning_rate': '0.0002994', 'epoch': '0.01685', 'train/total_time_seconds': '11.11', 'train/time_per_step_avg': '0.02102', 'train/epoch_time_elapsed': '55.72', 'train/estimated_remaining_minutes': '0.7407'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+{'eval_loss': '4.318', 'eval_runtime': '7.874', 'eval_samples_per_second': '1210', 'eval_steps_per_second': '1.27', 'epoch': '0.01685', 'train/total_time_seconds': '11.11', 'train/time_per_step_avg': '0.02102', 'train/epoch_time_elapsed': '63.59', 'train/estimated_remaining_minutes': '0.7407'}
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 145.98it/s]
+ 24%|██▍       | 600/2500 [01:08<01:28, 21.56it/s]?, ?it/s]
+{'loss': '4.301', 'grad_norm': '1.148', 'learning_rate': '0.0003', 'epoch': '0.01753', 'train/total_time_seconds': '11.54', 'train/time_per_step_avg': '0.02119', 'train/epoch_time_elapsed': '64.56', 'train/estimated_remaining_minutes': '0.7326'}
+{'loss': '4.226', 'grad_norm': '1.305', 'learning_rate': '0.0003', 'epoch': '0.0182', 'train/total_time_seconds': '11.97', 'train/time_per_step_avg': '0.02125', 'train/epoch_time_elapsed': '65.48', 'train/estimated_remaining_minutes': '0.7241'}
+{'loss': '4.174', 'grad_norm': '1.133', 'learning_rate': '0.0003', 'epoch': '0.01887', 'train/total_time_seconds': '12.39', 'train/time_per_step_avg': '0.02123', 'train/epoch_time_elapsed': '66.4', 'train/estimated_remaining_minutes': '0.7152'}
+{'loss': '4.122', 'grad_norm': '0.9727', 'learning_rate': '0.0003', 'epoch': '0.01955', 'train/total_time_seconds': '12.81', 'train/time_per_step_avg': '0.0211', 'train/epoch_time_elapsed': '67.31', 'train/estimated_remaining_minutes': '0.7065'}
+{'loss': '4.078', 'grad_norm': '1.211', 'learning_rate': '0.0003', 'epoch': '0.02022', 'train/total_time_seconds': '13.22', 'train/time_per_step_avg': '0.02113', 'train/epoch_time_elapsed': '68.23', 'train/estimated_remaining_minutes': '0.6979'}
+  0%|          | 0/10 [00:00<?, ?it/s]

zain/Activation/wandb/run-20260818_154123-mv5l58cq/logs/debug-internal.log

@@ -11,3 +11,7 @@
 {"time":"2026-08-18T15:41:39.107833948Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
 {"time":"2026-08-18T15:41:53.838786585Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":7,"history_lines":9,"events_offset":1,"events_lines":2,"console_offset":13,"console_lines":1}
 {"time":"2026-08-18T15:41:54.083266227Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-18T15:42:08.838687556Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":16,"history_lines":7,"events_offset":3,"events_lines":2,"console_offset":15,"console_lines":27}
+{"time":"2026-08-18T15:42:09.345507091Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-18T15:42:23.838894979Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":23,"history_lines":6,"events_offset":5,"events_lines":2,"console_offset":35,"console_lines":1}
+{"time":"2026-08-18T15:42:24.163808856Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}