Auto upload zain 2026-08-12T23:05:39.337686 (part 3)

zain/Activation/wandb/debug-internal.log

@@ -47,3 +47,45 @@
 {"time":"2026-08-12T23:00:12.23343263Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
 {"time":"2026-08-12T23:00:27.077426313Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":45,"history_lines":3,"events_offset":38,"events_lines":2,"console_offset":84,"console_lines":10}
 {"time":"2026-08-12T23:00:27.199618213Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:00:42.078828946Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":48,"history_lines":2,"events_offset":40,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-12T23:00:42.245309073Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:00:57.078070331Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":50,"history_lines":3,"events_offset":42,"events_lines":2,"console_offset":94,"console_lines":10}
+{"time":"2026-08-12T23:00:57.204086609Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:01:12.077201676Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":53,"history_lines":2,"events_offset":44,"events_lines":2,"console_offset":100,"console_lines":1}
+{"time":"2026-08-12T23:01:12.198006429Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:01:27.077924623Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":55,"history_lines":2,"events_offset":46,"events_lines":2,"console_offset":104,"console_lines":9}
+{"time":"2026-08-12T23:01:27.182682516Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:01:42.077961512Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":57,"history_lines":3,"events_offset":48,"events_lines":2,"console_offset":110,"console_lines":1}
+{"time":"2026-08-12T23:01:42.213811875Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:01:57.077777951Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":60,"history_lines":2,"events_offset":50,"events_lines":2,"console_offset":113,"console_lines":10}
+{"time":"2026-08-12T23:01:57.221530401Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:02:12.078021149Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":62,"history_lines":3,"events_offset":52,"events_lines":2,"console_offset":120,"console_lines":1}
+{"time":"2026-08-12T23:02:12.200523347Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:02:27.077228027Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":65,"history_lines":2,"events_offset":54,"events_lines":2,"console_offset":123,"console_lines":10}
+{"time":"2026-08-12T23:02:27.213269343Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:02:42.077949475Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":67,"history_lines":2,"events_offset":56,"events_lines":2,"console_offset":130,"console_lines":1}
+{"time":"2026-08-12T23:02:42.193118425Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:02:57.077286213Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":69,"history_lines":3,"events_offset":58,"events_lines":2,"console_offset":130,"console_lines":1}
+{"time":"2026-08-12T23:02:57.197768762Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:03:12.077825137Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":72,"history_lines":2,"events_offset":60,"events_lines":2,"console_offset":133,"console_lines":12}
+{"time":"2026-08-12T23:03:12.199228892Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:03:27.077537793Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":74,"history_lines":3,"events_offset":62,"events_lines":2,"console_offset":140,"console_lines":1}
+{"time":"2026-08-12T23:03:27.230629296Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:03:42.078034294Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":77,"history_lines":2,"events_offset":64,"events_lines":2,"console_offset":145,"console_lines":10}
+{"time":"2026-08-12T23:03:42.193085801Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:03:57.077383264Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":79,"history_lines":3,"events_offset":66,"events_lines":2,"console_offset":150,"console_lines":1}
+{"time":"2026-08-12T23:03:57.202324297Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:04:12.077898302Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":82,"history_lines":2,"events_offset":68,"events_lines":2,"console_offset":155,"console_lines":10}
+{"time":"2026-08-12T23:04:12.232728468Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:04:27.077816532Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":84,"history_lines":2,"events_offset":70,"events_lines":2,"console_offset":160,"console_lines":1}
+{"time":"2026-08-12T23:04:27.223742246Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:04:42.077397434Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":86,"history_lines":3,"events_offset":72,"events_lines":2,"console_offset":165,"console_lines":10}
+{"time":"2026-08-12T23:04:42.184847332Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:04:57.078035499Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":89,"history_lines":2,"events_offset":74,"events_lines":2,"console_offset":170,"console_lines":1}
+{"time":"2026-08-12T23:04:57.180764382Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:05:12.077941859Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":91,"history_lines":3,"events_offset":76,"events_lines":2,"console_offset":175,"console_lines":10}
+{"time":"2026-08-12T23:05:12.189439572Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:05:27.078134737Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":94,"history_lines":2,"events_offset":78,"events_lines":2,"console_offset":180,"console_lines":1}
+{"time":"2026-08-12T23:05:27.239753585Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:05:42.077996453Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":96,"history_lines":2,"events_offset":80,"events_lines":2,"console_offset":185,"console_lines":9}
+{"time":"2026-08-12T23:05:42.188706417Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}

zain/Activation/wandb/run-20260812_225525-hf77resg/files/output.log

@@ -88,8 +88,107 @@ Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 16.
   - 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, 16.76it/s]
- 40%|███▉      | 997/2500 [05:11<07:43,  3.24it/s]?, ?it/s]
+ 40%|████      | 1000/2500 [05:12<07:40,  3.26it/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': '2.66', 'grad_norm': '0.4355', 'learning_rate': '0.0007', 'epoch': '0.06202', 'train/total_time_seconds': '238', 'train/time_per_step_avg': '0.2603', 'train/epoch_time_elapsed': '286.7', 'train/estimated_remaining_minutes': '6.813'}
 {'loss': '2.647', 'grad_norm': '0.4551', 'learning_rate': '0.0007', 'epoch': '0.06336', 'train/total_time_seconds': '243.3', 'train/time_per_step_avg': '0.2613', 'train/epoch_time_elapsed': '292.9', 'train/estimated_remaining_minutes': '6.729'}
 {'loss': '2.623', 'grad_norm': '0.4551', 'learning_rate': '0.0007', 'epoch': '0.06471', 'train/total_time_seconds': '248.3', 'train/time_per_step_avg': '0.2583', 'train/epoch_time_elapsed': '299.1', 'train/estimated_remaining_minutes': '6.639'}
 {'loss': '2.61', 'grad_norm': '0.4629', 'learning_rate': '0.0007', 'epoch': '0.06606', 'train/total_time_seconds': '253.4', 'train/time_per_step_avg': '0.2579', 'train/epoch_time_elapsed': '305.2', 'train/estimated_remaining_minutes': '6.552'}
+{'loss': '2.611', 'grad_norm': '0.4883', 'learning_rate': '0.0007', 'epoch': '0.06741', 'train/total_time_seconds': '258.7', 'train/time_per_step_avg': '0.2583', 'train/epoch_time_elapsed': '311.4', 'train/estimated_remaining_minutes': '6.466'}
+  - 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, 17.59it/s]
+ 44%|████▍     | 1100/2500 [05:43<07:08,  3.27it/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': '2.582', 'grad_norm': '0.4238', 'learning_rate': '0.0007', 'epoch': '0.06876', 'train/total_time_seconds': '263.8', 'train/time_per_step_avg': '0.2576', 'train/epoch_time_elapsed': '317.8', 'train/estimated_remaining_minutes': '6.379'}
+{'loss': '2.568', 'grad_norm': '0.4453', 'learning_rate': '0.0007', 'epoch': '0.0701', 'train/total_time_seconds': '269', 'train/time_per_step_avg': '0.257', 'train/epoch_time_elapsed': '324', 'train/estimated_remaining_minutes': '6.293'}
+{'loss': '2.555', 'grad_norm': '0.4473', 'learning_rate': '0.0007', 'epoch': '0.07145', 'train/total_time_seconds': '274.1', 'train/time_per_step_avg': '0.2577', 'train/epoch_time_elapsed': '330.1', 'train/estimated_remaining_minutes': '6.206'}
+{'loss': '2.537', 'grad_norm': '0.4453', 'learning_rate': '0.0007', 'epoch': '0.0728', 'train/total_time_seconds': '279.3', 'train/time_per_step_avg': '0.2584', 'train/epoch_time_elapsed': '336.3', 'train/estimated_remaining_minutes': '6.12'}
+{'loss': '2.534', 'grad_norm': '0.4316', 'learning_rate': '0.0007', 'epoch': '0.07415', 'train/total_time_seconds': '284.4', 'train/time_per_step_avg': '0.2575', 'train/epoch_time_elapsed': '342.4', 'train/estimated_remaining_minutes': '6.033'}
+  - 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, 16.13it/s]
+ 48%|████▊     | 1200/2500 [06:14<06:37,  3.27it/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': '2.499', 'grad_norm': '0.4277', 'learning_rate': '0.0007', 'epoch': '0.0755', 'train/total_time_seconds': '289.7', 'train/time_per_step_avg': '0.2596', 'train/epoch_time_elapsed': '349', 'train/estimated_remaining_minutes': '5.95'}
+{'loss': '2.489', 'grad_norm': '0.4219', 'learning_rate': '0.0007', 'epoch': '0.07685', 'train/total_time_seconds': '294.9', 'train/time_per_step_avg': '0.2592', 'train/epoch_time_elapsed': '355.1', 'train/estimated_remaining_minutes': '5.863'}
+{'loss': '2.489', 'grad_norm': '0.4199', 'learning_rate': '0.0007', 'epoch': '0.07819', 'train/total_time_seconds': '300', 'train/time_per_step_avg': '0.2592', 'train/epoch_time_elapsed': '361.3', 'train/estimated_remaining_minutes': '5.776'}
+{'loss': '2.485', 'grad_norm': '0.4141', 'learning_rate': '0.0007', 'epoch': '0.07954', 'train/total_time_seconds': '305.1', 'train/time_per_step_avg': '0.2587', 'train/epoch_time_elapsed': '367.4', 'train/estimated_remaining_minutes': '5.689'}
+{'loss': '2.477', 'grad_norm': '0.5117', 'learning_rate': '0.0007', 'epoch': '0.08089', 'train/total_time_seconds': '310.3', 'train/time_per_step_avg': '0.2591', 'train/epoch_time_elapsed': '373.6', 'train/estimated_remaining_minutes': '5.603'}
+  - 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, 13.72it/s]
+ 52%|█████▏    | 1300/2500 [06:45<06:18,  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': '2.464', 'grad_norm': '0.418', 'learning_rate': '0.0007', 'epoch': '0.08224', 'train/total_time_seconds': '315.5', 'train/time_per_step_avg': '0.2573', 'train/epoch_time_elapsed': '380.1', 'train/estimated_remaining_minutes': '5.516'}
+{'loss': '2.451', 'grad_norm': '0.3906', 'learning_rate': '0.0007', 'epoch': '0.08359', 'train/total_time_seconds': '320.6', 'train/time_per_step_avg': '0.2569', 'train/epoch_time_elapsed': '386.2', 'train/estimated_remaining_minutes': '5.429'}
+{'loss': '2.441', 'grad_norm': '0.4453', 'learning_rate': '0.0007', 'epoch': '0.08493', 'train/total_time_seconds': '325.7', 'train/time_per_step_avg': '0.2571', 'train/epoch_time_elapsed': '392.4', 'train/estimated_remaining_minutes': '5.343'}
+{'loss': '2.423', 'grad_norm': '0.418', 'learning_rate': '0.0007', 'epoch': '0.08628', 'train/total_time_seconds': '330.9', 'train/time_per_step_avg': '0.2574', 'train/epoch_time_elapsed': '398.5', 'train/estimated_remaining_minutes': '5.256'}
+{'loss': '2.415', 'grad_norm': '0.4453', 'learning_rate': '0.0007', 'epoch': '0.08763', 'train/total_time_seconds': '336.1', 'train/time_per_step_avg': '0.2575', 'train/epoch_time_elapsed': '404.7', 'train/estimated_remaining_minutes': '5.17'}
+  - 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, 15.73it/s]
+ 56%|█████▌    | 1400/2500 [07:16<05:35,  3.28it/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': '2.408', 'grad_norm': '0.4102', 'learning_rate': '0.0007', 'epoch': '0.08898', 'train/total_time_seconds': '341.3', 'train/time_per_step_avg': '0.2582', 'train/epoch_time_elapsed': '411.2', 'train/estimated_remaining_minutes': '5.085'}
+{'loss': '2.397', 'grad_norm': '0.4121', 'learning_rate': '0.0007', 'epoch': '0.09033', 'train/total_time_seconds': '346.4', 'train/time_per_step_avg': '0.2584', 'train/epoch_time_elapsed': '417.3', 'train/estimated_remaining_minutes': '4.998'}
+{'loss': '2.399', 'grad_norm': '0.4316', 'learning_rate': '0.0007', 'epoch': '0.09168', 'train/total_time_seconds': '351.6', 'train/time_per_step_avg': '0.259', 'train/epoch_time_elapsed': '423.6', 'train/estimated_remaining_minutes': '4.912'}
+{'loss': '2.377', 'grad_norm': '0.4023', 'learning_rate': '0.0007', 'epoch': '0.09302', 'train/total_time_seconds': '356.8', 'train/time_per_step_avg': '0.2588', 'train/epoch_time_elapsed': '429.7', 'train/estimated_remaining_minutes': '4.826'}
+{'loss': '2.37', 'grad_norm': '0.4062', 'learning_rate': '0.0007', 'epoch': '0.09437', 'train/total_time_seconds': '361.9', 'train/time_per_step_avg': '0.2584', 'train/epoch_time_elapsed': '435.8', 'train/estimated_remaining_minutes': '4.739'}
+  - 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, 17.13it/s]
+ 60%|██████    | 1500/2500 [07:47<05:05,  3.27it/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': '2.381', 'grad_norm': '0.4219', 'learning_rate': '0.0007', 'epoch': '0.09572', 'train/total_time_seconds': '367.3', 'train/time_per_step_avg': '0.2598', 'train/epoch_time_elapsed': '442.4', 'train/estimated_remaining_minutes': '4.655'}
+{'loss': '2.359', 'grad_norm': '0.4316', 'learning_rate': '0.0007', 'epoch': '0.09707', 'train/total_time_seconds': '372.4', 'train/time_per_step_avg': '0.2596', 'train/epoch_time_elapsed': '448.5', 'train/estimated_remaining_minutes': '4.569'}
+{'loss': '2.348', 'grad_norm': '0.4707', 'learning_rate': '0.0007', 'epoch': '0.09842', 'train/total_time_seconds': '377.6', 'train/time_per_step_avg': '0.2592', 'train/epoch_time_elapsed': '454.7', 'train/estimated_remaining_minutes': '4.482'}
+{'loss': '2.336', 'grad_norm': '0.4004', 'learning_rate': '0.0007', 'epoch': '0.09976', 'train/total_time_seconds': '382.7', 'train/time_per_step_avg': '0.2593', 'train/epoch_time_elapsed': '460.9', 'train/estimated_remaining_minutes': '4.396'}
+{'loss': '2.323', 'grad_norm': '0.4141', 'learning_rate': '0.0007', 'epoch': '0.1011', 'train/total_time_seconds': '387.9', 'train/time_per_step_avg': '0.2598', 'train/epoch_time_elapsed': '467.1', 'train/estimated_remaining_minutes': '4.31'}
+  - 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, 15.71it/s]
+ 64%|██████▍   | 1600/2500 [08:18<04:34,  3.28it/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': '2.327', 'grad_norm': '0.4375', 'learning_rate': '0.0007', 'epoch': '0.1025', 'train/total_time_seconds': '393', 'train/time_per_step_avg': '0.2574', 'train/epoch_time_elapsed': '473.5', 'train/estimated_remaining_minutes': '4.223'}
+{'loss': '2.315', 'grad_norm': '0.4082', 'learning_rate': '0.0007', 'epoch': '0.1038', 'train/total_time_seconds': '398.1', 'train/time_per_step_avg': '0.2575', 'train/epoch_time_elapsed': '479.6', 'train/estimated_remaining_minutes': '4.136'}
+{'loss': '2.302', 'grad_norm': '0.4082', 'learning_rate': '0.0007', 'epoch': '0.1052', 'train/total_time_seconds': '403.3', 'train/time_per_step_avg': '0.2579', 'train/epoch_time_elapsed': '485.8', 'train/estimated_remaining_minutes': '4.051'}
+{'loss': '2.324', 'grad_norm': '0.4082', 'learning_rate': '0.0007', 'epoch': '0.1065', 'train/total_time_seconds': '408.5', 'train/time_per_step_avg': '0.2582', 'train/epoch_time_elapsed': '492', 'train/estimated_remaining_minutes': '3.965'}
+{'loss': '2.289', 'grad_norm': '0.4004', 'learning_rate': '0.0007', 'epoch': '0.1079', 'train/total_time_seconds': '413.6', 'train/time_per_step_avg': '0.2576', 'train/epoch_time_elapsed': '498.1', 'train/estimated_remaining_minutes': '3.878'}
+  - 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, 16.40it/s]
+ 68%|██████▊   | 1700/2500 [08:49<04:04,  3.27it/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': '2.288', 'grad_norm': '0.4141', 'learning_rate': '0.0007', 'epoch': '0.1092', 'train/total_time_seconds': '418.8', 'train/time_per_step_avg': '0.2576', 'train/epoch_time_elapsed': '504.5', 'train/estimated_remaining_minutes': '3.791'}
+{'loss': '2.301', 'grad_norm': '0.416', 'learning_rate': '0.0007', 'epoch': '0.1105', 'train/total_time_seconds': '423.9', 'train/time_per_step_avg': '0.258', 'train/epoch_time_elapsed': '510.7', 'train/estimated_remaining_minutes': '3.705'}
+{'loss': '2.285', 'grad_norm': '0.3809', 'learning_rate': '0.0007', 'epoch': '0.1119', 'train/total_time_seconds': '429.1', 'train/time_per_step_avg': '0.2571', 'train/epoch_time_elapsed': '516.8', 'train/estimated_remaining_minutes': '3.619'}
+{'loss': '2.268', 'grad_norm': '0.4082', 'learning_rate': '0.0007', 'epoch': '0.1132', 'train/total_time_seconds': '434.2', 'train/time_per_step_avg': '0.2569', 'train/epoch_time_elapsed': '523', 'train/estimated_remaining_minutes': '3.532'}
+{'loss': '2.28', 'grad_norm': '0.4238', 'learning_rate': '0.0007', 'epoch': '0.1146', 'train/total_time_seconds': '439.3', 'train/time_per_step_avg': '0.2569', 'train/epoch_time_elapsed': '529.1', 'train/estimated_remaining_minutes': '3.446'}
+  - 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, 15.74it/s]
+ 72%|███████▏  | 1800/2500 [09:20<03:33,  3.27it/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': '2.258', 'grad_norm': '0.4375', 'learning_rate': '0.0007', 'epoch': '0.1159', 'train/total_time_seconds': '444.7', 'train/time_per_step_avg': '0.259', 'train/epoch_time_elapsed': '535.7', 'train/estimated_remaining_minutes': '3.361'}
+{'loss': '2.247', 'grad_norm': '0.3945', 'learning_rate': '0.0007', 'epoch': '0.1173', 'train/total_time_seconds': '449.8', 'train/time_per_step_avg': '0.259', 'train/epoch_time_elapsed': '541.9', 'train/estimated_remaining_minutes': '3.275'}
+{'loss': '2.252', 'grad_norm': '0.4199', 'learning_rate': '0.0007', 'epoch': '0.1186', 'train/total_time_seconds': '454.9', 'train/time_per_step_avg': '0.2587', 'train/epoch_time_elapsed': '548', 'train/estimated_remaining_minutes': '3.188'}
+{'loss': '2.265', 'grad_norm': '0.3867', 'learning_rate': '0.0007', 'epoch': '0.12', 'train/total_time_seconds': '460', 'train/time_per_step_avg': '0.2582', 'train/epoch_time_elapsed': '554', 'train/estimated_remaining_minutes': '3.101'}
+{'loss': '2.261', 'grad_norm': '0.4082', 'learning_rate': '0.0007', 'epoch': '0.1213', 'train/total_time_seconds': '465.2', 'train/time_per_step_avg': '0.2582', 'train/epoch_time_elapsed': '560.2', 'train/estimated_remaining_minutes': '3.015'}
+  - 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, 15.58it/s]
+ 76%|███████▌  | 1900/2500 [09:51<03:05,  3.24it/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': '2.254', 'grad_norm': '0.4102', 'learning_rate': '0.0007', 'epoch': '0.1227', 'train/total_time_seconds': '470.3', 'train/time_per_step_avg': '0.2559', 'train/epoch_time_elapsed': '566.6', 'train/estimated_remaining_minutes': '2.928'}
+{'loss': '2.221', 'grad_norm': '0.4043', 'learning_rate': '0.0007', 'epoch': '0.124', 'train/total_time_seconds': '475.3', 'train/time_per_step_avg': '0.2552', 'train/epoch_time_elapsed': '572.7', 'train/estimated_remaining_minutes': '2.842'}
+{'loss': '2.222', 'grad_norm': '0.377', 'learning_rate': '0.0007', 'epoch': '0.1254', 'train/total_time_seconds': '480.5', 'train/time_per_step_avg': '0.2563', 'train/epoch_time_elapsed': '578.9', 'train/estimated_remaining_minutes': '2.756'}
+{'loss': '2.214', 'grad_norm': '0.3828', 'learning_rate': '0.0007', 'epoch': '0.1267', 'train/total_time_seconds': '485.7', 'train/time_per_step_avg': '0.2571', 'train/epoch_time_elapsed': '585.1', 'train/estimated_remaining_minutes': '2.67'}
+{'loss': '2.221', 'grad_norm': '0.3945', 'learning_rate': '0.0007', 'epoch': '0.1281', 'train/total_time_seconds': '490.9', 'train/time_per_step_avg': '0.2573', 'train/epoch_time_elapsed': '591.3', 'train/estimated_remaining_minutes': '2.584'}
+  - 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, 16.30it/s]
+ 79%|███████▉  | 1978/2500 [10:16<02:42,  3.20it/s], ?it/s]
+{'loss': '2.199', 'grad_norm': '0.3945', 'learning_rate': '0.0007', 'epoch': '0.1294', 'train/total_time_seconds': '496', 'train/time_per_step_avg': '0.2576', 'train/epoch_time_elapsed': '597.7', 'train/estimated_remaining_minutes': '2.497'}
+{'loss': '2.21', 'grad_norm': '0.4102', 'learning_rate': '0.0007', 'epoch': '0.1308', 'train/total_time_seconds': '501.2', 'train/time_per_step_avg': '0.2586', 'train/epoch_time_elapsed': '603.8', 'train/estimated_remaining_minutes': '2.411'}
+{'loss': '2.185', 'grad_norm': '0.3965', 'learning_rate': '0.0007', 'epoch': '0.1321', 'train/total_time_seconds': '506.4', 'train/time_per_step_avg': '0.2581', 'train/epoch_time_elapsed': '610', 'train/estimated_remaining_minutes': '2.325'}

zain/Activation/wandb/run-20260812_225525-hf77resg/logs/debug-internal.log

@@ -47,3 +47,45 @@
 {"time":"2026-08-12T23:00:12.23343263Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
 {"time":"2026-08-12T23:00:27.077426313Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":45,"history_lines":3,"events_offset":38,"events_lines":2,"console_offset":84,"console_lines":10}
 {"time":"2026-08-12T23:00:27.199618213Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:00:42.078828946Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":48,"history_lines":2,"events_offset":40,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-12T23:00:42.245309073Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:00:57.078070331Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":50,"history_lines":3,"events_offset":42,"events_lines":2,"console_offset":94,"console_lines":10}
+{"time":"2026-08-12T23:00:57.204086609Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:01:12.077201676Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":53,"history_lines":2,"events_offset":44,"events_lines":2,"console_offset":100,"console_lines":1}
+{"time":"2026-08-12T23:01:12.198006429Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:01:27.077924623Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":55,"history_lines":2,"events_offset":46,"events_lines":2,"console_offset":104,"console_lines":9}
+{"time":"2026-08-12T23:01:27.182682516Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:01:42.077961512Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":57,"history_lines":3,"events_offset":48,"events_lines":2,"console_offset":110,"console_lines":1}
+{"time":"2026-08-12T23:01:42.213811875Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:01:57.077777951Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":60,"history_lines":2,"events_offset":50,"events_lines":2,"console_offset":113,"console_lines":10}
+{"time":"2026-08-12T23:01:57.221530401Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:02:12.078021149Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":62,"history_lines":3,"events_offset":52,"events_lines":2,"console_offset":120,"console_lines":1}
+{"time":"2026-08-12T23:02:12.200523347Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:02:27.077228027Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":65,"history_lines":2,"events_offset":54,"events_lines":2,"console_offset":123,"console_lines":10}
+{"time":"2026-08-12T23:02:27.213269343Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:02:42.077949475Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":67,"history_lines":2,"events_offset":56,"events_lines":2,"console_offset":130,"console_lines":1}
+{"time":"2026-08-12T23:02:42.193118425Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:02:57.077286213Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":69,"history_lines":3,"events_offset":58,"events_lines":2,"console_offset":130,"console_lines":1}
+{"time":"2026-08-12T23:02:57.197768762Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:03:12.077825137Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":72,"history_lines":2,"events_offset":60,"events_lines":2,"console_offset":133,"console_lines":12}
+{"time":"2026-08-12T23:03:12.199228892Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:03:27.077537793Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":74,"history_lines":3,"events_offset":62,"events_lines":2,"console_offset":140,"console_lines":1}
+{"time":"2026-08-12T23:03:27.230629296Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:03:42.078034294Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":77,"history_lines":2,"events_offset":64,"events_lines":2,"console_offset":145,"console_lines":10}
+{"time":"2026-08-12T23:03:42.193085801Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:03:57.077383264Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":79,"history_lines":3,"events_offset":66,"events_lines":2,"console_offset":150,"console_lines":1}
+{"time":"2026-08-12T23:03:57.202324297Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:04:12.077898302Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":82,"history_lines":2,"events_offset":68,"events_lines":2,"console_offset":155,"console_lines":10}
+{"time":"2026-08-12T23:04:12.232728468Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:04:27.077816532Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":84,"history_lines":2,"events_offset":70,"events_lines":2,"console_offset":160,"console_lines":1}
+{"time":"2026-08-12T23:04:27.223742246Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:04:42.077397434Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":86,"history_lines":3,"events_offset":72,"events_lines":2,"console_offset":165,"console_lines":10}
+{"time":"2026-08-12T23:04:42.184847332Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:04:57.078035499Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":89,"history_lines":2,"events_offset":74,"events_lines":2,"console_offset":170,"console_lines":1}
+{"time":"2026-08-12T23:04:57.180764382Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:05:12.077941859Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":91,"history_lines":3,"events_offset":76,"events_lines":2,"console_offset":175,"console_lines":10}
+{"time":"2026-08-12T23:05:12.189439572Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:05:27.078134737Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":94,"history_lines":2,"events_offset":78,"events_lines":2,"console_offset":180,"console_lines":1}
+{"time":"2026-08-12T23:05:27.239753585Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T23:05:42.077996453Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":96,"history_lines":2,"events_offset":80,"events_lines":2,"console_offset":185,"console_lines":9}
+{"time":"2026-08-12T23:05:42.188706417Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}

zain/Activation/wandb/run-20260812_225525-hf77resg/run-hf77resg.wandb

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5df2846f51aea2fd207e766ab7eafcbcc53ff5bdbfd66bc73858cb44195ffe18
-size 327680
+oid sha256:f045fadabadc51069de364d13fa35e63232304a0524e461adce7d39cbf490813
+size 655360