Auto upload zain 2026-08-12T22:50:19.164217 (part 3)

zain/Activation/wandb/debug-internal.log

@@ -39,3 +39,43 @@
 {"time":"2026-08-12T22:44:57.806264315Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
 {"time":"2026-08-12T22:45:12.66281333Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":35,"history_lines":2,"events_offset":30,"events_lines":2,"console_offset":63,"console_lines":10}
 {"time":"2026-08-12T22:45:12.810852756Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:45:27.662584141Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":37,"history_lines":2,"events_offset":32,"events_lines":2,"console_offset":70,"console_lines":1}
+{"time":"2026-08-12T22:45:27.794784745Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:45:42.663258489Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":39,"history_lines":3,"events_offset":34,"events_lines":2,"console_offset":70,"console_lines":1}
+{"time":"2026-08-12T22:45:42.850852204Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:45:57.662937624Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":42,"history_lines":2,"events_offset":36,"events_lines":2,"console_offset":73,"console_lines":12}
+{"time":"2026-08-12T22:45:57.80693389Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:46:12.662804753Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":44,"history_lines":2,"events_offset":38,"events_lines":2,"console_offset":80,"console_lines":1}
+{"time":"2026-08-12T22:46:12.798005405Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:46:27.662629099Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":46,"history_lines":3,"events_offset":40,"events_lines":2,"console_offset":85,"console_lines":10}
+{"time":"2026-08-12T22:46:27.77060706Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:46:42.662834862Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":49,"history_lines":2,"events_offset":42,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-12T22:46:42.771984837Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:46:57.662457151Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":51,"history_lines":2,"events_offset":44,"events_lines":2,"console_offset":95,"console_lines":9}
+{"time":"2026-08-12T22:46:57.781154126Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:47:12.662782505Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":53,"history_lines":3,"events_offset":46,"events_lines":2,"console_offset":100,"console_lines":1}
+{"time":"2026-08-12T22:47:12.801530339Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:47:27.662583093Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":56,"history_lines":2,"events_offset":48,"events_lines":2,"console_offset":104,"console_lines":10}
+{"time":"2026-08-12T22:47:27.768889846Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:47:42.663327169Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":58,"history_lines":2,"events_offset":50,"events_lines":2,"console_offset":110,"console_lines":1}
+{"time":"2026-08-12T22:47:42.785619124Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:47:57.663243976Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":60,"history_lines":3,"events_offset":52,"events_lines":2,"console_offset":114,"console_lines":10}
+{"time":"2026-08-12T22:47:57.820043037Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:48:12.66334227Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":63,"history_lines":2,"events_offset":54,"events_lines":2,"console_offset":120,"console_lines":1}
+{"time":"2026-08-12T22:48:12.805122197Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:48:27.663076872Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":65,"history_lines":2,"events_offset":56,"events_lines":2,"console_offset":124,"console_lines":9}
+{"time":"2026-08-12T22:48:27.792216461Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:48:42.663340794Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":67,"history_lines":3,"events_offset":58,"events_lines":2,"console_offset":130,"console_lines":1}
+{"time":"2026-08-12T22:48:42.900721656Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:48:57.662763507Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":70,"history_lines":2,"events_offset":60,"events_lines":2,"console_offset":133,"console_lines":10}
+{"time":"2026-08-12T22:48:57.802049179Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:49:12.662978517Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":72,"history_lines":2,"events_offset":62,"events_lines":2,"console_offset":140,"console_lines":1}
+{"time":"2026-08-12T22:49:12.778403666Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:49:27.663287472Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":74,"history_lines":3,"events_offset":64,"events_lines":2,"console_offset":140,"console_lines":1}
+{"time":"2026-08-12T22:49:27.78207664Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:49:42.662884197Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":77,"history_lines":2,"events_offset":66,"events_lines":2,"console_offset":143,"console_lines":12}
+{"time":"2026-08-12T22:49:42.805635061Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:49:57.663318835Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":79,"history_lines":2,"events_offset":68,"events_lines":2,"console_offset":150,"console_lines":1}
+{"time":"2026-08-12T22:49:57.792796554Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:50:12.663041496Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":81,"history_lines":3,"events_offset":70,"events_lines":2,"console_offset":155,"console_lines":10}
+{"time":"2026-08-12T22:50:12.852358868Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}

zain/Activation/wandb/run-20260812_224111-g6mesp08/files/output.log

@@ -68,7 +68,104 @@ Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 13.
   - 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, 14.19it/s]
- 31%|███       | 766/2500 [04:06<09:08,  3.16it/s]?, ?it/s]
+ 32%|███▏      | 800/2500 [04:17<08:59,  3.15it/s][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': '3.012', 'grad_norm': '0.3945', 'learning_rate': '0.0007', 'epoch': '0.04853', 'train/total_time_seconds': '193.4', 'train/time_per_step_avg': '0.2674', 'train/epoch_time_elapsed': '230.9', 'train/estimated_remaining_minutes': '7.968'}
 {'loss': '2.973', 'grad_norm': '0.4062', 'learning_rate': '0.0007', 'epoch': '0.04988', 'train/total_time_seconds': '198.7', 'train/time_per_step_avg': '0.2674', 'train/epoch_time_elapsed': '237.2', 'train/estimated_remaining_minutes': '7.877'}
 {'loss': '2.957', 'grad_norm': '0.4102', 'learning_rate': '0.0007', 'epoch': '0.05123', 'train/total_time_seconds': '204.1', 'train/time_per_step_avg': '0.2678', 'train/epoch_time_elapsed': '243.6', 'train/estimated_remaining_minutes': '7.787'}
+{'loss': '2.959', 'grad_norm': '0.4277', 'learning_rate': '0.0007', 'epoch': '0.05258', 'train/total_time_seconds': '209.6', 'train/time_per_step_avg': '0.2692', 'train/epoch_time_elapsed': '250.1', 'train/estimated_remaining_minutes': '7.704'}
+{'loss': '2.907', 'grad_norm': '0.4297', 'learning_rate': '0.0007', 'epoch': '0.05393', 'train/total_time_seconds': '215', 'train/time_per_step_avg': '0.2696', 'train/epoch_time_elapsed': '256.5', 'train/estimated_remaining_minutes': '7.614'}
+  - 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
+  - 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, 14.83it/s]
+ 36%|███▌      | 900/2500 [04:49<08:28,  3.15it/s][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': '2.894', 'grad_norm': '0.4355', 'learning_rate': '0.0007', 'epoch': '0.05527', 'train/total_time_seconds': '220.3', 'train/time_per_step_avg': '0.2697', 'train/epoch_time_elapsed': '263.1', 'train/estimated_remaining_minutes': '7.524'}
+{'loss': '2.863', 'grad_norm': '0.4297', 'learning_rate': '0.0007', 'epoch': '0.05662', 'train/total_time_seconds': '225.7', 'train/time_per_step_avg': '0.2697', 'train/epoch_time_elapsed': '269.4', 'train/estimated_remaining_minutes': '7.433'}
+{'loss': '2.819', 'grad_norm': '0.3887', 'learning_rate': '0.0007', 'epoch': '0.05797', 'train/total_time_seconds': '231', 'train/time_per_step_avg': '0.2697', 'train/epoch_time_elapsed': '275.8', 'train/estimated_remaining_minutes': '7.343'}
+{'loss': '2.794', 'grad_norm': '0.3848', 'learning_rate': '0.0007', 'epoch': '0.05932', 'train/total_time_seconds': '236.4', 'train/time_per_step_avg': '0.2677', 'train/epoch_time_elapsed': '282.1', 'train/estimated_remaining_minutes': '7.252'}
+{'loss': '2.78', 'grad_norm': '0.4023', 'learning_rate': '0.0007', 'epoch': '0.06067', 'train/total_time_seconds': '241.7', 'train/time_per_step_avg': '0.2672', 'train/epoch_time_elapsed': '288.4', 'train/estimated_remaining_minutes': '7.162'}
+  - 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
+  - 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, 15.83it/s]
+ 40%|████      | 1000/2500 [05:21<08:00,  3.12it/s][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': '2.752', 'grad_norm': '0.416', 'learning_rate': '0.0007', 'epoch': '0.06202', 'train/total_time_seconds': '247', 'train/time_per_step_avg': '0.2668', 'train/epoch_time_elapsed': '295', 'train/estimated_remaining_minutes': '7.071'}
+{'loss': '2.739', 'grad_norm': '0.4473', 'learning_rate': '0.0007', 'epoch': '0.06336', 'train/total_time_seconds': '252.4', 'train/time_per_step_avg': '0.2668', 'train/epoch_time_elapsed': '301.3', 'train/estimated_remaining_minutes': '6.98'}
+{'loss': '2.711', 'grad_norm': '0.4414', 'learning_rate': '0.0007', 'epoch': '0.06471', 'train/total_time_seconds': '257.7', 'train/time_per_step_avg': '0.2666', 'train/epoch_time_elapsed': '307.6', 'train/estimated_remaining_minutes': '6.89'}
+{'loss': '2.692', 'grad_norm': '0.3945', 'learning_rate': '0.0007', 'epoch': '0.06606', 'train/total_time_seconds': '263.3', 'train/time_per_step_avg': '0.2693', 'train/epoch_time_elapsed': '314.2', 'train/estimated_remaining_minutes': '6.807'}
+{'loss': '2.691', 'grad_norm': '0.3633', 'learning_rate': '0.0007', 'epoch': '0.06741', 'train/total_time_seconds': '268.7', 'train/time_per_step_avg': '0.2696', 'train/epoch_time_elapsed': '320.6', 'train/estimated_remaining_minutes': '6.717'}
+  - 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
+  - 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, 14.40it/s]
+ 44%|████▍     | 1100/2500 [05:53<07:27,  3.13it/s][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': '2.659', 'grad_norm': '0.3711', 'learning_rate': '0.0007', 'epoch': '0.06876', 'train/total_time_seconds': '274', 'train/time_per_step_avg': '0.2702', 'train/epoch_time_elapsed': '327.2', 'train/estimated_remaining_minutes': '6.627'}
+{'loss': '2.644', 'grad_norm': '0.4062', 'learning_rate': '0.0007', 'epoch': '0.0701', 'train/total_time_seconds': '279.4', 'train/time_per_step_avg': '0.27', 'train/epoch_time_elapsed': '333.5', 'train/estimated_remaining_minutes': '6.537'}
+{'loss': '2.629', 'grad_norm': '0.3906', 'learning_rate': '0.0007', 'epoch': '0.07145', 'train/total_time_seconds': '284.8', 'train/time_per_step_avg': '0.2704', 'train/epoch_time_elapsed': '339.9', 'train/estimated_remaining_minutes': '6.447'}
+{'loss': '2.607', 'grad_norm': '0.3789', 'learning_rate': '0.0007', 'epoch': '0.0728', 'train/total_time_seconds': '290.1', 'train/time_per_step_avg': '0.268', 'train/epoch_time_elapsed': '346.3', 'train/estimated_remaining_minutes': '6.357'}
+{'loss': '2.6', 'grad_norm': '0.4121', 'learning_rate': '0.0007', 'epoch': '0.07415', 'train/total_time_seconds': '295.5', 'train/time_per_step_avg': '0.2683', 'train/epoch_time_elapsed': '352.6', 'train/estimated_remaining_minutes': '6.268'}
+  - 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
+  - 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, 11.71it/s]
+ 48%|████▊     | 1200/2500 [06:25<06:48,  3.18it/s][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': '2.566', 'grad_norm': '0.3848', 'learning_rate': '0.0007', 'epoch': '0.0755', 'train/total_time_seconds': '300.8', 'train/time_per_step_avg': '0.268', 'train/epoch_time_elapsed': '359.3', 'train/estimated_remaining_minutes': '6.178'}
+{'loss': '2.554', 'grad_norm': '0.418', 'learning_rate': '0.0007', 'epoch': '0.07685', 'train/total_time_seconds': '306.2', 'train/time_per_step_avg': '0.2685', 'train/epoch_time_elapsed': '365.7', 'train/estimated_remaining_minutes': '6.089'}
+{'loss': '2.55', 'grad_norm': '0.3789', 'learning_rate': '0.0007', 'epoch': '0.07819', 'train/total_time_seconds': '311.8', 'train/time_per_step_avg': '0.2706', 'train/epoch_time_elapsed': '372.3', 'train/estimated_remaining_minutes': '6.003'}
+{'loss': '2.543', 'grad_norm': '0.3867', 'learning_rate': '0.0007', 'epoch': '0.07954', 'train/total_time_seconds': '317.2', 'train/time_per_step_avg': '0.2705', 'train/epoch_time_elapsed': '378.6', 'train/estimated_remaining_minutes': '5.913'}
+{'loss': '2.534', 'grad_norm': '0.457', 'learning_rate': '0.0007', 'epoch': '0.08089', 'train/total_time_seconds': '322.5', 'train/time_per_step_avg': '0.2698', 'train/epoch_time_elapsed': '384.9', 'train/estimated_remaining_minutes': '5.823'}
+  - 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
+  - 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, 13.14it/s]
+ 52%|█████▏    | 1300/2500 [06:57<06:19,  3.16it/s][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': '2.519', 'grad_norm': '0.3594', 'learning_rate': '0.0007', 'epoch': '0.08224', 'train/total_time_seconds': '327.8', 'train/time_per_step_avg': '0.2697', 'train/epoch_time_elapsed': '391.5', 'train/estimated_remaining_minutes': '5.732'}
+{'loss': '2.505', 'grad_norm': '0.3633', 'learning_rate': '0.0007', 'epoch': '0.08359', 'train/total_time_seconds': '333.1', 'train/time_per_step_avg': '0.269', 'train/epoch_time_elapsed': '397.8', 'train/estimated_remaining_minutes': '5.642'}
+{'loss': '2.49', 'grad_norm': '0.3672', 'learning_rate': '0.0007', 'epoch': '0.08493', 'train/total_time_seconds': '338.5', 'train/time_per_step_avg': '0.2664', 'train/epoch_time_elapsed': '404.1', 'train/estimated_remaining_minutes': '5.551'}
+{'loss': '2.47', 'grad_norm': '0.3691', 'learning_rate': '0.0007', 'epoch': '0.08628', 'train/total_time_seconds': '343.8', 'train/time_per_step_avg': '0.2665', 'train/epoch_time_elapsed': '410.5', 'train/estimated_remaining_minutes': '5.462'}
+{'loss': '2.462', 'grad_norm': '0.3691', 'learning_rate': '0.0007', 'epoch': '0.08763', 'train/total_time_seconds': '349.2', 'train/time_per_step_avg': '0.267', 'train/epoch_time_elapsed': '416.8', 'train/estimated_remaining_minutes': '5.372'}
+  - 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
+  - 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, 16.05it/s]
+ 56%|█████▌    | 1400/2500 [07:30<05:46,  3.17it/s][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': '2.454', 'grad_norm': '0.3613', 'learning_rate': '0.0007', 'epoch': '0.08898', 'train/total_time_seconds': '354.5', 'train/time_per_step_avg': '0.2671', 'train/epoch_time_elapsed': '423.5', 'train/estimated_remaining_minutes': '5.282'}
+{'loss': '2.442', 'grad_norm': '0.3652', 'learning_rate': '0.0007', 'epoch': '0.09033', 'train/total_time_seconds': '359.9', 'train/time_per_step_avg': '0.2674', 'train/epoch_time_elapsed': '429.8', 'train/estimated_remaining_minutes': '5.192'}
+{'loss': '2.441', 'grad_norm': '0.4082', 'learning_rate': '0.0007', 'epoch': '0.09168', 'train/total_time_seconds': '365.4', 'train/time_per_step_avg': '0.2696', 'train/epoch_time_elapsed': '436.3', 'train/estimated_remaining_minutes': '5.105'}
+{'loss': '2.418', 'grad_norm': '0.3691', 'learning_rate': '0.0007', 'epoch': '0.09302', 'train/total_time_seconds': '370.8', 'train/time_per_step_avg': '0.2696', 'train/epoch_time_elapsed': '442.7', 'train/estimated_remaining_minutes': '5.015'}
+{'loss': '2.408', 'grad_norm': '0.3477', 'learning_rate': '0.0007', 'epoch': '0.09437', 'train/total_time_seconds': '376.2', 'train/time_per_step_avg': '0.2699', 'train/epoch_time_elapsed': '449.1', 'train/estimated_remaining_minutes': '4.926'}
+  - 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
+  - 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, 14.22it/s]
+ 60%|██████    | 1500/2500 [08:01<05:17,  3.15it/s][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': '2.417', 'grad_norm': '0.3809', 'learning_rate': '0.0007', 'epoch': '0.09572', 'train/total_time_seconds': '381.5', 'train/time_per_step_avg': '0.2697', 'train/epoch_time_elapsed': '455.6', 'train/estimated_remaining_minutes': '4.836'}
+{'loss': '2.397', 'grad_norm': '0.4004', 'learning_rate': '0.0007', 'epoch': '0.09707', 'train/total_time_seconds': '386.8', 'train/time_per_step_avg': '0.2696', 'train/epoch_time_elapsed': '462', 'train/estimated_remaining_minutes': '4.746'}
+{'loss': '2.385', 'grad_norm': '0.3809', 'learning_rate': '0.0007', 'epoch': '0.09842', 'train/total_time_seconds': '392.1', 'train/time_per_step_avg': '0.2671', 'train/epoch_time_elapsed': '468.3', 'train/estimated_remaining_minutes': '4.655'}
+{'loss': '2.369', 'grad_norm': '0.3574', 'learning_rate': '0.0007', 'epoch': '0.09976', 'train/total_time_seconds': '397.5', 'train/time_per_step_avg': '0.2668', 'train/epoch_time_elapsed': '474.6', 'train/estimated_remaining_minutes': '4.565'}
+{'loss': '2.357', 'grad_norm': '0.3652', 'learning_rate': '0.0007', 'epoch': '0.1011', 'train/total_time_seconds': '402.8', 'train/time_per_step_avg': '0.2663', 'train/epoch_time_elapsed': '480.9', 'train/estimated_remaining_minutes': '4.475'}
+  - 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
+  - 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, 15.50it/s]
+ 64%|██████▍   | 1600/2500 [08:34<04:46,  3.14it/s][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': '2.359', 'grad_norm': '0.3594', 'learning_rate': '0.0007', 'epoch': '0.1025', 'train/total_time_seconds': '408.1', 'train/time_per_step_avg': '0.2666', 'train/epoch_time_elapsed': '487.5', 'train/estimated_remaining_minutes': '4.386'}
+{'loss': '2.347', 'grad_norm': '0.3984', 'learning_rate': '0.0007', 'epoch': '0.1038', 'train/total_time_seconds': '413.7', 'train/time_per_step_avg': '0.2687', 'train/epoch_time_elapsed': '494.1', 'train/estimated_remaining_minutes': '4.298'}
+{'loss': '2.337', 'grad_norm': '0.3691', 'learning_rate': '0.0007', 'epoch': '0.1052', 'train/total_time_seconds': '419.1', 'train/time_per_step_avg': '0.2696', 'train/epoch_time_elapsed': '500.5', 'train/estimated_remaining_minutes': '4.209'}
+{'loss': '2.355', 'grad_norm': '0.3477', 'learning_rate': '0.0007', 'epoch': '0.1065', 'train/total_time_seconds': '424.4', 'train/time_per_step_avg': '0.2697', 'train/epoch_time_elapsed': '506.8', 'train/estimated_remaining_minutes': '4.119'}
+{'loss': '2.322', 'grad_norm': '0.4062', 'learning_rate': '0.0007', 'epoch': '0.1079', 'train/total_time_seconds': '429.8', 'train/time_per_step_avg': '0.2699', 'train/epoch_time_elapsed': '513.2', 'train/estimated_remaining_minutes': '4.029'}
+  - 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
+  - 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, 14.62it/s]
+ 68%|██████▊   | 1700/2500 [09:05<04:12,  3.17it/s][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': '2.321', 'grad_norm': '0.3516', 'learning_rate': '0.0007', 'epoch': '0.1092', 'train/total_time_seconds': '435.1', 'train/time_per_step_avg': '0.2694', 'train/epoch_time_elapsed': '519.8', 'train/estimated_remaining_minutes': '3.939'}
+{'loss': '2.332', 'grad_norm': '0.3516', 'learning_rate': '0.0007', 'epoch': '0.1105', 'train/total_time_seconds': '440.4', 'train/time_per_step_avg': '0.2671', 'train/epoch_time_elapsed': '526', 'train/estimated_remaining_minutes': '3.849'}
+{'loss': '2.313', 'grad_norm': '0.3262', 'learning_rate': '0.0007', 'epoch': '0.1119', 'train/total_time_seconds': '445.8', 'train/time_per_step_avg': '0.267', 'train/epoch_time_elapsed': '532.4', 'train/estimated_remaining_minutes': '3.76'}
+{'loss': '2.296', 'grad_norm': '0.3418', 'learning_rate': '0.0007', 'epoch': '0.1132', 'train/total_time_seconds': '451.1', 'train/time_per_step_avg': '0.2669', 'train/epoch_time_elapsed': '538.7', 'train/estimated_remaining_minutes': '3.67'}
+{'loss': '2.308', 'grad_norm': '0.377', 'learning_rate': '0.0007', 'epoch': '0.1146', 'train/total_time_seconds': '456.4', 'train/time_per_step_avg': '0.2665', 'train/epoch_time_elapsed': '545', 'train/estimated_remaining_minutes': '3.58'}
+  - 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
+  - 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, 13.92it/s]
+ 69%|██████▊   | 1716/2500 [09:11<04:07,  3.17it/s], ?it/s]

zain/Activation/wandb/run-20260812_224111-g6mesp08/logs/debug-internal.log

@@ -39,3 +39,43 @@
 {"time":"2026-08-12T22:44:57.806264315Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
 {"time":"2026-08-12T22:45:12.66281333Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":35,"history_lines":2,"events_offset":30,"events_lines":2,"console_offset":63,"console_lines":10}
 {"time":"2026-08-12T22:45:12.810852756Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:45:27.662584141Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":37,"history_lines":2,"events_offset":32,"events_lines":2,"console_offset":70,"console_lines":1}
+{"time":"2026-08-12T22:45:27.794784745Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:45:42.663258489Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":39,"history_lines":3,"events_offset":34,"events_lines":2,"console_offset":70,"console_lines":1}
+{"time":"2026-08-12T22:45:42.850852204Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:45:57.662937624Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":42,"history_lines":2,"events_offset":36,"events_lines":2,"console_offset":73,"console_lines":12}
+{"time":"2026-08-12T22:45:57.80693389Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:46:12.662804753Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":44,"history_lines":2,"events_offset":38,"events_lines":2,"console_offset":80,"console_lines":1}
+{"time":"2026-08-12T22:46:12.798005405Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:46:27.662629099Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":46,"history_lines":3,"events_offset":40,"events_lines":2,"console_offset":85,"console_lines":10}
+{"time":"2026-08-12T22:46:27.77060706Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:46:42.662834862Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":49,"history_lines":2,"events_offset":42,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-12T22:46:42.771984837Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:46:57.662457151Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":51,"history_lines":2,"events_offset":44,"events_lines":2,"console_offset":95,"console_lines":9}
+{"time":"2026-08-12T22:46:57.781154126Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:47:12.662782505Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":53,"history_lines":3,"events_offset":46,"events_lines":2,"console_offset":100,"console_lines":1}
+{"time":"2026-08-12T22:47:12.801530339Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:47:27.662583093Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":56,"history_lines":2,"events_offset":48,"events_lines":2,"console_offset":104,"console_lines":10}
+{"time":"2026-08-12T22:47:27.768889846Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:47:42.663327169Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":58,"history_lines":2,"events_offset":50,"events_lines":2,"console_offset":110,"console_lines":1}
+{"time":"2026-08-12T22:47:42.785619124Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:47:57.663243976Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":60,"history_lines":3,"events_offset":52,"events_lines":2,"console_offset":114,"console_lines":10}
+{"time":"2026-08-12T22:47:57.820043037Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:48:12.66334227Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":63,"history_lines":2,"events_offset":54,"events_lines":2,"console_offset":120,"console_lines":1}
+{"time":"2026-08-12T22:48:12.805122197Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:48:27.663076872Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":65,"history_lines":2,"events_offset":56,"events_lines":2,"console_offset":124,"console_lines":9}
+{"time":"2026-08-12T22:48:27.792216461Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:48:42.663340794Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":67,"history_lines":3,"events_offset":58,"events_lines":2,"console_offset":130,"console_lines":1}
+{"time":"2026-08-12T22:48:42.900721656Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:48:57.662763507Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":70,"history_lines":2,"events_offset":60,"events_lines":2,"console_offset":133,"console_lines":10}
+{"time":"2026-08-12T22:48:57.802049179Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:49:12.662978517Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":72,"history_lines":2,"events_offset":62,"events_lines":2,"console_offset":140,"console_lines":1}
+{"time":"2026-08-12T22:49:12.778403666Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:49:27.663287472Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":74,"history_lines":3,"events_offset":64,"events_lines":2,"console_offset":140,"console_lines":1}
+{"time":"2026-08-12T22:49:27.78207664Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:49:42.662884197Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":77,"history_lines":2,"events_offset":66,"events_lines":2,"console_offset":143,"console_lines":12}
+{"time":"2026-08-12T22:49:42.805635061Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:49:57.663318835Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":79,"history_lines":2,"events_offset":68,"events_lines":2,"console_offset":150,"console_lines":1}
+{"time":"2026-08-12T22:49:57.792796554Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T22:50:12.663041496Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":81,"history_lines":3,"events_offset":70,"events_lines":2,"console_offset":155,"console_lines":10}
+{"time":"2026-08-12T22:50:12.852358868Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}

zain/Activation/wandb/run-20260812_224111-g6mesp08/run-g6mesp08.wandb

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9e3697394f44baf30175c8c2f84ceee8ba4bb1f6d40e8044d211332b7f114694
-size 229376
+oid sha256:06b4b4fea40d339e946481c5ba93364948dfa7874a1ce338d616d3dcc40a5e91
+size 589824