Auto upload zain 2026-08-18T15:43:05.347262 (part 3)

.gitattributes

@@ -41,3 +41,4 @@ zain/Activation/wandb/run-20260818_151642-lf0kv584/run-lf0kv584.wandb filter=lfs
 zain/Activation/wandb/run-20260818_152324-7wl5fa0w/run-7wl5fa0w.wandb filter=lfs diff=lfs merge=lfs -text
 zain/Activation/wandb/run-20260818_153050-d5bw8p9o/run-d5bw8p9o.wandb filter=lfs diff=lfs merge=lfs -text
 zain/Activation/wandb/run-20260818_153817-b8i366dc/run-b8i366dc.wandb filter=lfs diff=lfs merge=lfs -text
+zain/Activation/wandb/run-20260818_154123-mv5l58cq/run-mv5l58cq.wandb filter=lfs diff=lfs merge=lfs -text

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

@@ -55,10 +55,40 @@ Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 158
   - 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]
+ 24%|██▍       | 600/2500 [01:16<01:28, 21.56it[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': '4.301', '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]
+  - 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': '4.058', 'eval_runtime': '7.915', 'eval_samples_per_second': '1204', 'eval_steps_per_second': '1.263', 'epoch': '0.02022', 'train/total_time_seconds': '13.22', 'train/time_per_step_avg': '0.02113', 'train/epoch_time_elapsed': '76.15', 'train/estimated_remaining_minutes': '0.6979'}
+  - 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, 142.69it/s]
+ 28%|██▊       | 700/2500 [01:28<01:27, 20.63it[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': '4.02', 'grad_norm': '1.234', 'learning_rate': '0.0003', 'epoch': '0.0209', 'train/total_time_seconds': '13.64', 'train/time_per_step_avg': '0.02092', 'train/epoch_time_elapsed': '77.09', 'train/estimated_remaining_minutes': '0.6892'}
+{'loss': '3.995', 'grad_norm': '1.422', 'learning_rate': '0.0003', 'epoch': '0.02157', 'train/total_time_seconds': '14.05', 'train/time_per_step_avg': '0.0208', 'train/epoch_time_elapsed': '78', 'train/estimated_remaining_minutes': '0.6805'}
+{'loss': '3.962', 'grad_norm': '1.188', 'learning_rate': '0.0003', 'epoch': '0.02224', 'train/total_time_seconds': '14.48', 'train/time_per_step_avg': '0.0209', 'train/epoch_time_elapsed': '78.92', 'train/estimated_remaining_minutes': '0.6727'}
+{'loss': '3.922', 'grad_norm': '1.258', 'learning_rate': '0.0003', 'epoch': '0.02292', 'train/total_time_seconds': '14.93', 'train/time_per_step_avg': '0.02123', 'train/epoch_time_elapsed': '79.92', 'train/estimated_remaining_minutes': '0.6659'}
+{'loss': '3.915', 'grad_norm': '1.344', 'learning_rate': '0.0003', 'epoch': '0.02359', 'train/total_time_seconds': '15.4', 'train/time_per_step_avg': '0.02175', 'train/epoch_time_elapsed': '80.89', 'train/estimated_remaining_minutes': '0.6599'}
+  - 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': '3.9', 'eval_runtime': '7.943', 'eval_samples_per_second': '1199', 'eval_steps_per_second': '1.259', 'epoch': '0.02359', 'train/total_time_seconds': '15.4', 'train/time_per_step_avg': '0.02175', 'train/epoch_time_elapsed': '88.84', 'train/estimated_remaining_minutes': '0.6599'}
+  - 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, 164.77it/s]
+ 32%|███▏      | 800/2500 [01:41<01:16, 22.31it[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': '3.883', 'grad_norm': '1.266', 'learning_rate': '0.0003', 'epoch': '0.02427', 'train/total_time_seconds': '15.86', 'train/time_per_step_avg': '0.02224', 'train/epoch_time_elapsed': '89.84', 'train/estimated_remaining_minutes': '0.6535'}
+{'loss': '3.874', 'grad_norm': '1.273', 'learning_rate': '0.0003', 'epoch': '0.02494', 'train/total_time_seconds': '16.31', 'train/time_per_step_avg': '0.02261', 'train/epoch_time_elapsed': '90.82', 'train/estimated_remaining_minutes': '0.6465'}
+{'loss': '3.829', 'grad_norm': '1.5', 'learning_rate': '0.0003', 'epoch': '0.02562', 'train/total_time_seconds': '16.72', 'train/time_per_step_avg': '0.02243', 'train/epoch_time_elapsed': '91.72', 'train/estimated_remaining_minutes': '0.638'}
+{'loss': '3.79', 'grad_norm': '1.266', 'learning_rate': '0.0003', 'epoch': '0.02629', 'train/total_time_seconds': '17.13', 'train/time_per_step_avg': '0.02201', 'train/epoch_time_elapsed': '92.61', 'train/estimated_remaining_minutes': '0.6296'}
+{'loss': '3.799', 'grad_norm': '1.102', 'learning_rate': '0.0003', 'epoch': '0.02696', 'train/total_time_seconds': '17.54', 'train/time_per_step_avg': '0.02143', 'train/epoch_time_elapsed': '93.51', 'train/estimated_remaining_minutes': '0.6212'}
+  - 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': '3.794', 'eval_runtime': '7.917', 'eval_samples_per_second': '1203', 'eval_steps_per_second': '1.263', 'epoch': '0.02696', 'train/total_time_seconds': '17.54', 'train/time_per_step_avg': '0.02143', 'train/epoch_time_elapsed': '101.4', 'train/estimated_remaining_minutes': '0.6212'}
+  - 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, 155.59it/s]
+ 35%|███▌      | 875/2500 [01:44<01:14, 21.92it/s]?, ?it/s]
+{'loss': '3.761', 'grad_norm': '1.141', 'learning_rate': '0.0003', 'epoch': '0.02764', 'train/total_time_seconds': '17.96', 'train/time_per_step_avg': '0.02102', 'train/epoch_time_elapsed': '102.4', 'train/estimated_remaining_minutes': '0.6133'}
+{'loss': '3.754', 'grad_norm': '1.234', 'learning_rate': '0.0003', 'epoch': '0.02831', 'train/total_time_seconds': '18.38', 'train/time_per_step_avg': '0.02074', 'train/epoch_time_elapsed': '103.3', 'train/estimated_remaining_minutes': '0.6055'}
+{'loss': '3.781', 'grad_norm': '1.367', 'learning_rate': '0.0003', 'epoch': '0.02899', 'train/total_time_seconds': '18.82', 'train/time_per_step_avg': '0.021', 'train/epoch_time_elapsed': '104.3', 'train/estimated_remaining_minutes': '0.5982'}

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

@@ -15,3 +15,9 @@
 {"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"}
+{"time":"2026-08-18T15:42:38.838208182Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":29,"history_lines":6,"events_offset":7,"events_lines":2,"console_offset":41,"console_lines":23}
+{"time":"2026-08-18T15:42:39.163065229Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-18T15:42:53.83793847Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":35,"history_lines":8,"events_offset":9,"events_lines":2,"console_offset":57,"console_lines":1}
+{"time":"2026-08-18T15:42:54.287863999Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-18T15:43:08.838036308Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":43,"history_lines":8,"events_offset":11,"events_lines":2,"console_offset":63,"console_lines":31}
+{"time":"2026-08-18T15:43:09.143645213Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}