Auto upload zain 2026-08-14T23:07:46.851605 (part 8)

.gitattributes

@@ -63,3 +63,10 @@ zain/Activation/wandb/run-20260814_212208-n6fjpg9o/run-n6fjpg9o.wandb filter=lfs
 zain/Activation/out/glu-waleed10-100L_trash_run/training_log.jsonl filter=lfs diff=lfs merge=lfs -text
 zain/Activation/wandb/run-20260814_212835-gqskne0x/run-gqskne0x.wandb filter=lfs diff=lfs merge=lfs -text
 zain/Activation/out/glu-silu-waleed10-100L_trash_run/training_log.jsonl filter=lfs diff=lfs merge=lfs -text
+zain/Activation/wandb/run-20260814_213153-q9vcmje5/run-q9vcmje5.wandb filter=lfs diff=lfs merge=lfs -text
+zain/Activation/wandb/run-20260814_214549-k24y90ws/run-k24y90ws.wandb filter=lfs diff=lfs merge=lfs -text
+zain/Activation/wandb/run-20260814_215902-npap5k3g/run-npap5k3g.wandb filter=lfs diff=lfs merge=lfs -text
+zain/Activation/wandb/run-20260814_221154-a3607q94/run-a3607q94.wandb filter=lfs diff=lfs merge=lfs -text
+zain/Activation/wandb/run-20260814_222456-wcvy5vyw/run-wcvy5vyw.wandb filter=lfs diff=lfs merge=lfs -text
+zain/Activation/wandb/run-20260814_223829-uuv4h936/run-uuv4h936.wandb filter=lfs diff=lfs merge=lfs -text
+zain/Activation/wandb/run-20260814_225218-x87hwrtd/run-x87hwrtd.wandb filter=lfs diff=lfs merge=lfs -text

zain/Activation/wandb/run-20260814_212835-gqskne0x/logs/debug-core.log

@@ -75,3 +75,11 @@
 {"time":"2026-08-14T21:28:35.344777056Z","level":"INFO","msg":"handleInformInit: received","streamId":"gqskne0x","id":"3(@)"}
 {"time":"2026-08-14T21:28:35.603024157Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"gqskne0x","id":"3(@)"}
 {"time":"2026-08-14T21:28:41.078893683Z","level":"INFO","msg":"connection: cancelling request","id":"3(@)","requestId":"yxbo55z35lse"}
+{"time":"2026-08-14T21:35:05.007052341Z","level":"INFO","msg":"connection: cancelling request","id":"3(@)","requestId":"yxbo55z35lse"}
+{"time":"2026-08-14T21:35:06.745707472Z","level":"INFO","msg":"connection: cancelling request","id":"3(@)","requestId":"yxbo55z35lse"}
+{"time":"2026-08-14T21:35:06.767299957Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"gqskne0x","id":"3(@)"}
+{"time":"2026-08-14T21:35:06.773662133Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"gqskne0x","id":"3(@)"}
+{"time":"2026-08-14T21:35:08.806158062Z","level":"INFO","msg":"connection: closing","id":"3(@)"}
+{"time":"2026-08-14T21:35:08.806241813Z","level":"INFO","msg":"connection: closed successfully","id":"3(@)"}
+{"time":"2026-08-14T21:35:08.806161779Z","level":"INFO","msg":"processOutgoingData: finished","id":"3(@)"}
+{"time":"2026-08-14T21:35:08.806251477Z","level":"INFO","msg":"connection: ManageConnectionData: connection closed","id":"3(@)"}

zain/Activation/wandb/run-20260814_212835-gqskne0x/logs/debug-internal.log

@@ -45,3 +45,49 @@
 {"time":"2026-08-14T21:29:52.226871571Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
 {"time":"2026-08-14T21:30:06.110393089Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":11,"history_lines":2,"events_offset":9,"events_lines":2,"console_offset":15,"console_lines":10}
 {"time":"2026-08-14T21:30:07.570174219Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:30:21.109617614Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":13,"history_lines":3,"events_offset":11,"events_lines":2,"console_offset":21,"console_lines":1}
+{"time":"2026-08-14T21:30:22.080315159Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:30:36.115961679Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":16,"history_lines":3,"events_offset":13,"events_lines":2,"console_offset":25,"console_lines":11}
+{"time":"2026-08-14T21:30:37.217389873Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:30:51.100718712Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":19,"history_lines":2,"events_offset":15,"events_lines":2,"console_offset":31,"console_lines":1}
+{"time":"2026-08-14T21:30:51.948020311Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:31:06.117731609Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":21,"history_lines":3,"events_offset":17,"events_lines":2,"console_offset":36,"console_lines":10}
+{"time":"2026-08-14T21:31:07.223434557Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:31:21.099230907Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":24,"history_lines":2,"events_offset":19,"events_lines":2,"console_offset":41,"console_lines":1}
+{"time":"2026-08-14T21:31:21.759032284Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:31:36.114372704Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":26,"history_lines":3,"events_offset":21,"events_lines":2,"console_offset":46,"console_lines":10}
+{"time":"2026-08-14T21:31:37.153517827Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:31:51.098719406Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":29,"history_lines":2,"events_offset":23,"events_lines":2,"console_offset":51,"console_lines":1}
+{"time":"2026-08-14T21:31:52.120465709Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:32:06.108268856Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":31,"history_lines":2,"events_offset":25,"events_lines":2,"console_offset":56,"console_lines":9}
+{"time":"2026-08-14T21:32:07.025091977Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:32:21.114007643Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":33,"history_lines":2,"events_offset":27,"events_lines":2,"console_offset":61,"console_lines":1}
+{"time":"2026-08-14T21:32:22.043468825Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:32:36.110666778Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":35,"history_lines":2,"events_offset":29,"events_lines":2,"console_offset":65,"console_lines":9}
+{"time":"2026-08-14T21:32:36.96385382Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:32:51.106950797Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":37,"history_lines":2,"events_offset":31,"events_lines":2,"console_offset":71,"console_lines":1}
+{"time":"2026-08-14T21:32:52.043139409Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:33:06.101277485Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":39,"history_lines":1,"events_offset":33,"events_lines":2,"console_offset":71,"console_lines":1}
+{"time":"2026-08-14T21:33:07.00621301Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:33:21.115155817Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":40,"history_lines":2,"events_offset":35,"events_lines":2,"console_offset":74,"console_lines":10}
+{"time":"2026-08-14T21:33:22.004559582Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:33:36.113278989Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":42,"history_lines":2,"events_offset":37,"events_lines":2,"console_offset":81,"console_lines":1}
+{"time":"2026-08-14T21:33:37.036173708Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:33:51.103713137Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":44,"history_lines":2,"events_offset":39,"events_lines":2,"console_offset":81,"console_lines":1}
+{"time":"2026-08-14T21:33:51.923559316Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:34:06.111807693Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":46,"history_lines":1,"events_offset":41,"events_lines":2,"console_offset":84,"console_lines":10}
+{"time":"2026-08-14T21:34:07.118394557Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:34:21.110439909Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":47,"history_lines":2,"events_offset":43,"events_lines":2,"console_offset":91,"console_lines":1}
+{"time":"2026-08-14T21:34:22.046475792Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:34:36.099853771Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":49,"history_lines":1,"events_offset":45,"events_lines":2,"console_offset":91,"console_lines":1}
+{"time":"2026-08-14T21:34:36.908754879Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:34:51.101796618Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":50,"history_lines":2,"events_offset":47,"events_lines":2,"console_offset":91,"console_lines":1}
+{"time":"2026-08-14T21:34:51.852581483Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:35:05.824230118Z","level":"INFO","msg":"fileTransfer: Close: file transfer manager closed"}
+{"time":"2026-08-14T21:35:05.833974131Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":52,"history_lines":1,"events_offset":49,"events_lines":1,"console_offset":94,"console_lines":17,"uploaded_len":3,"complete":true,"exit_code":0}
+{"time":"2026-08-14T21:35:06.721587751Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:35:06.722695731Z","level":"INFO","msg":"handler: operation stats","stats":{}}
+{"time":"2026-08-14T21:35:06.767338139Z","level":"INFO","msg":"stream: finishing up"}
+{"time":"2026-08-14T21:35:06.767370539Z","level":"INFO","msg":"handler: closed"}
+{"time":"2026-08-14T21:35:06.767476774Z","level":"INFO","msg":"sender: closed"}
+{"time":"2026-08-14T21:35:06.767482411Z","level":"INFO","msg":"stream: all finished"}

zain/Activation/wandb/run-20260814_212835-gqskne0x/logs/debug.log

@@ -18,3 +18,8 @@ config: {'_wandb': {}}
 2026-08-14 21:28:36,078 INFO    MainThread:936813 [wandb_run.py:_config_callback():1346] config_cb None None {'transformers_version': '5.16.0.dev0', 'architectures': None, 'output_hidden_states': False, 'return_dict': True, 'dtype': None, 'chunk_size_feed_forward': 0, 'is_encoder_decoder': False, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'problem_type': None, 'vocab_size': 4096, 'hidden_size': 128, 'intermediate_size': 256, 'num_hidden_layers': 100, 'num_attention_heads': 4, 'num_key_value_heads': 4, 'hidden_act': 'silu', 'max_position_embeddings': 512, 'initializer_range': 0.02, 'rms_norm_eps': 1e-06, 'use_cache': False, 'pad_token_id': 0, 'bos_token_id': 1, 'eos_token_id': 2, 'pretraining_tp': 1, 'tie_word_embeddings': True, 'rope_parameters': {'rope_theta': 10000.0, 'rope_type': 'default'}, 'attention_bias': False, 'attention_dropout': 0.0, 'mlp_bias': False, 'head_dim': 32, '_name_or_path': '', 'tokenizer_name': 'w-ahmad/tiny-stories-tokenizer', 'mlp_type': 'glu', 'activation': 'silu-waleed10', 'waleed_beta': 4, 'model_type': 'tiny_llama', 'output_attentions': False, 'output_dir': 'out/glu-silu-waleed10-100L_trash_run', 'per_device_train_batch_size': 64, 'num_train_epochs': 1, 'max_steps': 3500, 'learning_rate': 9e-05, 'lr_scheduler_type': 'constant_with_warmup', 'lr_scheduler_kwargs': None, 'warmup_steps': 50, 'optim': 'adamw_torch_fused', 'optim_args': None, 'weight_decay': 0.0, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'optim_target_modules': None, 'gradient_accumulation_steps': 1, 'average_tokens_across_devices': True, 'max_grad_norm': 1.0, 'label_smoothing_factor': 0.0, 'bf16': True, 'fp16': False, 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'gradient_checkpointing': False, 'gradient_checkpointing_kwargs': None, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'use_liger_kernel': False, 'liger_kernel_config': None, 'neftune_noise_alpha': None, 'torch_empty_cache_steps': None, 'auto_find_batch_size': False, 'logging_strategy': 'steps', 'logging_steps': 20, 'logging_first_step': False, 'log_on_each_node': True, 'logging_nan_inf_filter': True, 'include_num_input_tokens_seen': 'no', 'log_level': 'passive', 'log_level_replica': 'warning', 'disable_tqdm': False, 'report_to': ['wandb'], 'run_name': 'LM-glu-silu-waleed10-100L-16.9M-20260814-212834', 'project': 'huggingface', 'trackio_space_id': None, 'trackio_bucket_id': None, 'trackio_static_space_id': None, 'eval_strategy': 'steps', 'eval_steps': 2498, 'eval_delay': 0, 'per_device_eval_batch_size': 1024, 'prediction_loss_only': False, 'eval_on_start': False, 'eval_do_concat_batches': True, 'eval_use_gather_object': False, 'eval_accumulation_steps': None, 'include_for_metrics': [], 'batch_eval_metrics': False, 'save_only_model': False, 'save_strategy': 'steps', 'save_steps': 50, 'save_on_each_node': False, 'save_total_limit': None, 'enable_jit_checkpoint': False, 'push_to_hub': False, 'hub_token': '<HUB_TOKEN>', 'hub_private_repo': None, 'hub_model_id': 'w-ahmad/6L-glu-silu-waleed10-100L', 'hub_strategy': 'every_save', 'hub_always_push': False, 'hub_revision': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'restore_callback_states_from_checkpoint': False, 'full_determinism': False, 'seed': 42, 'data_seed': 42, 'use_cpu': False, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}, 'parallelism_config': None, 'dataloader_drop_last': False, 'dataloader_num_workers': 0, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'dataloader_prefetch_factor': None, 'dataloader_multiprocessing_context': None, 'dataloader_in_order': True, 'remove_unused_columns': False, 'label_names': None, 'train_sampling_strategy': 'random', 'length_column_name': 'length', 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'ddp_static_graph': None, 'ddp_backend': None, 'ddp_timeout': 1800, 'fsdp': None, 'fsdp_config': None, 'deepspeed': None, 'debug': [], 'skip_memory_metrics': True, 'do_train': False, 'do_eval': True, 'do_predict': False, 'resume_from_checkpoint': None, 'local_rank': -1}
 2026-08-14 21:28:36,083 INFO    MainThread:936813 [wandb_config.py:__setitem__():155] [no run ID] config set model/num_parameters = 16934016 - <bound method Run._config_callback of <wandb.sdk.wandb_run.Run object at 0x14ee399dda90>>
 2026-08-14 21:28:36,083 INFO    MainThread:936813 [wandb_run.py:_config_callback():1346] config_cb model/num_parameters 16934016 None
+2026-08-14 21:35:05,006 INFO    MainThread:936813 [wandb_run.py:_finish():2383] finishing run deepnevro-deepnevro/research7/gqskne0x
+2026-08-14 21:35:05,006 INFO    MainThread:936813 [wandb_run.py:_atexit_cleanup():2588] got exitcode: 0
+2026-08-14 21:35:05,006 INFO    MainThread:936813 [wandb_run.py:_restore():2570] restore
+2026-08-14 21:35:05,006 INFO    MainThread:936813 [wandb_run.py:_restore():2576] restore done
+2026-08-14 21:35:06,766 INFO    MainThread:936813 [wandb_run.py:_footer_sync_info():3993] logging synced files

zain/Activation/wandb/run-20260814_212835-gqskne0x/run-gqskne0x.wandb

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:642685d115792bd6ab89f27f6b3b3941e2759dd1c0b094ae917478acfdb1e3ac
-size 14843904
+oid sha256:59d9ff1dae5584c3328fb08df040f9affb6b10bd792c5ca40be4072c95d32f90
+size 51423131

zain/Activation/wandb/run-20260814_213153-q9vcmje5/files/config.yaml

@@ -0,0 +1,439 @@
+_name_or_path:
+    value: ""
+_wandb:
+    value:
+        cli_version: 0.28.1
+        e:
+            avd7wjdsk6ji6qbb7ojy8wwedk5izmcm:
+                args:
+                    - --config
+                    - /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/configs/baseline.yaml
+                    - --variants
+                    - glu-waleed10
+                    - glu-silu-waleed10
+                    - mlp-waleed10
+                    - mlp-silu-waleed10
+                    - glu-waleed
+                    - glu-situglu_low
+                    - glu-waleedglu_low
+                codePath: sweep.py
+                codePathLocal: sweep.py
+                cpu_count: 112
+                cpu_count_logical: 224
+                cudaVersion: "12.4"
+                disk:
+                    /:
+                        total: "1560765693952"
+                        used: "709060804608"
+                email: deepnevro@gmail.com
+                executable: /mnt/data/zainulabideen/zain-exp/notebooks/my_env/bin/python
+                git:
+                    commit: 3fb47c6943d6927eafc39bd399a3d3a520bae74a
+                    remote: https://github.com/w-ahmad1a10/Activation.git
+                gpu: NVIDIA H100 80GB HBM3
+                gpu_count: 8
+                gpu_nvidia:
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-39c684a5-fde6-83d7-1663-0859795881ae
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-68012e5a-38b6-b643-0ca6-62fb66720bf3
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-132944c4-b689-2b5f-89a4-d730401677ab
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-2df386cc-6d26-d0e2-7a2d-a057b0d95864
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-bfa16575-1d94-1aa2-4537-2c93433f42ef
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-bc6c3e3c-9b90-09ca-c034-774961847c54
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-00a441e1-7c95-e7d6-4c35-43d6b291aea9
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-1c4d29a2-4647-6fce-d8fc-0c5ecfbbd6ea
+                host: deeplens-k3s-node1
+                memory:
+                    total: "2164089937920"
+                os: Linux-5.15.0-126-generic-x86_64-with-glibc2.35
+                program: /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/sweep.py
+                python: CPython 3.11.15
+                root: /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation
+                startedAt: "2026-08-14T21:31:53.799629Z"
+                writerId: avd7wjdsk6ji6qbb7ojy8wwedk5izmcm
+        m:
+            - "1": train/global_step
+              "6":
+                - 3
+              "7": []
+            - "2": '*'
+              "5": 1
+              "6":
+                - 1
+              "7": []
+        python_version: 3.11.15
+        t:
+            "1":
+                - 1
+                - 5
+                - 11
+                - 41
+                - 49
+                - 51
+                - 53
+                - 71
+            "2":
+                - 1
+                - 5
+                - 11
+                - 41
+                - 49
+                - 51
+                - 53
+                - 71
+            "3":
+                - 2
+                - 7
+                - 13
+                - 19
+                - 62
+                - 66
+            "4": 3.11.15
+            "5": 0.28.1
+            "6": 5.16.0.dev0
+            "9":
+                "1": transformers_trainer
+            "12": 0.28.1
+            "13": linux-x86_64
+accelerator_config:
+    value:
+        dispatch_batches: null
+        even_batches: true
+        gradient_accumulation_kwargs: null
+        non_blocking: false
+        split_batches: false
+        use_seedable_sampler: true
+activation:
+    value: waleed10
+adam_beta1:
+    value: 0.9
+adam_beta2:
+    value: 0.999
+adam_epsilon:
+    value: 1e-08
+architectures:
+    value: null
+attention_bias:
+    value: false
+attention_dropout:
+    value: 0
+auto_find_batch_size:
+    value: false
+average_tokens_across_devices:
+    value: true
+batch_eval_metrics:
+    value: false
+bf16:
+    value: true
+bf16_full_eval:
+    value: false
+bos_token_id:
+    value: 1
+chunk_size_feed_forward:
+    value: 0
+data_seed:
+    value: 42
+dataloader_drop_last:
+    value: false
+dataloader_in_order:
+    value: true
+dataloader_multiprocessing_context:
+    value: null
+dataloader_num_workers:
+    value: 0
+dataloader_persistent_workers:
+    value: false
+dataloader_pin_memory:
+    value: true
+dataloader_prefetch_factor:
+    value: null
+ddp_backend:
+    value: null
+ddp_broadcast_buffers:
+    value: null
+ddp_bucket_cap_mb:
+    value: null
+ddp_find_unused_parameters:
+    value: null
+ddp_static_graph:
+    value: null
+ddp_timeout:
+    value: 1800
+debug:
+    value: []
+deepspeed:
+    value: null
+disable_tqdm:
+    value: false
+do_eval:
+    value: true
+do_predict:
+    value: false
+do_train:
+    value: false
+dtype:
+    value: null
+enable_jit_checkpoint:
+    value: false
+eos_token_id:
+    value: 2
+eval_accumulation_steps:
+    value: null
+eval_delay:
+    value: 0
+eval_do_concat_batches:
+    value: true
+eval_on_start:
+    value: false
+eval_steps:
+    value: 60000
+eval_strategy:
+    value: steps
+eval_use_gather_object:
+    value: false
+fp16:
+    value: false
+fp16_full_eval:
+    value: false
+fsdp:
+    value: null
+fsdp_config:
+    value: null
+full_determinism:
+    value: false
+gradient_accumulation_steps:
+    value: 4
+gradient_checkpointing:
+    value: false
+gradient_checkpointing_kwargs:
+    value: null
+greater_is_better:
+    value: null
+head_dim:
+    value: 32
+hidden_act:
+    value: silu
+hidden_size:
+    value: 128
+hub_always_push:
+    value: false
+hub_model_id:
+    value: w-ahmad/A-glu-waleed10-21L
+hub_private_repo:
+    value: null
+hub_revision:
+    value: null
+hub_strategy:
+    value: every_save
+hub_token:
+    value: <HUB_TOKEN>
+id2label:
+    value:
+        "0": LABEL_0
+        "1": LABEL_1
+ignore_data_skip:
+    value: false
+include_for_metrics:
+    value: []
+include_num_input_tokens_seen:
+    value: "no"
+initializer_range:
+    value: 0.02
+intermediate_size:
+    value: 256
+is_encoder_decoder:
+    value: false
+label_names:
+    value: null
+label_smoothing_factor:
+    value: 0
+label2id:
+    value:
+        LABEL_0: 0
+        LABEL_1: 1
+learning_rate:
+    value: 0.001
+length_column_name:
+    value: length
+liger_kernel_config:
+    value: null
+load_best_model_at_end:
+    value: false
+local_rank:
+    value: -1
+log_level:
+    value: passive
+log_level_replica:
+    value: warning
+log_on_each_node:
+    value: true
+logging_first_step:
+    value: false
+logging_nan_inf_filter:
+    value: true
+logging_steps:
+    value: 20
+logging_strategy:
+    value: steps
+lr_scheduler_kwargs:
+    value: null
+lr_scheduler_type:
+    value: constant
+max_grad_norm:
+    value: 1
+max_position_embeddings:
+    value: 512
+max_steps:
+    value: 1000
+metric_for_best_model:
+    value: null
+mlp_bias:
+    value: false
+mlp_type:
+    value: glu
+model/num_parameters:
+    value: 3970432
+model_type:
+    value: tiny_llama
+neftune_noise_alpha:
+    value: null
+num_attention_heads:
+    value: 4
+num_hidden_layers:
+    value: 21
+num_key_value_heads:
+    value: 4
+num_train_epochs:
+    value: 1
+optim:
+    value: adamw_torch
+optim_args:
+    value: null
+optim_target_modules:
+    value: null
+output_attentions:
+    value: false
+output_dir:
+    value: out/glu-waleed10-21L_run
+output_hidden_states:
+    value: false
+pad_token_id:
+    value: 0
+parallelism_config:
+    value: null
+per_device_eval_batch_size:
+    value: 128
+per_device_train_batch_size:
+    value: 128
+prediction_loss_only:
+    value: false
+pretraining_tp:
+    value: 1
+problem_type:
+    value: null
+project:
+    value: huggingface
+push_to_hub:
+    value: false
+remove_unused_columns:
+    value: false
+report_to:
+    value:
+        - wandb
+restore_callback_states_from_checkpoint:
+    value: false
+resume_from_checkpoint:
+    value: null
+return_dict:
+    value: true
+rms_norm_eps:
+    value: 1e-06
+rope_parameters:
+    value:
+        rope_theta: 10000
+        rope_type: default
+run_name:
+    value: LM-glu-waleed10-21L-4.0M-20260814-213152
+save_on_each_node:
+    value: false
+save_only_model:
+    value: false
+save_steps:
+    value: 100
+save_strategy:
+    value: steps
+save_total_limit:
+    value: null
+seed:
+    value: 42
+skip_memory_metrics:
+    value: true
+tf32:
+    value: null
+tie_word_embeddings:
+    value: true
+tokenizer_name:
+    value: w-ahmad/tiny-stories-tokenizer
+torch_compile:
+    value: false
+torch_compile_backend:
+    value: null
+torch_compile_mode:
+    value: null
+torch_empty_cache_steps:
+    value: null
+trackio_bucket_id:
+    value: null
+trackio_space_id:
+    value: null
+trackio_static_space_id:
+    value: null
+train_sampling_strategy:
+    value: random
+transformers_version:
+    value: 5.16.0.dev0
+use_cache:
+    value: false
+use_cpu:
+    value: false
+use_liger_kernel:
+    value: false
+vocab_size:
+    value: 4096
+waleed_beta:
+    value: 10
+warmup_steps:
+    value: 0
+weight_decay:
+    value: 0.01

zain/Activation/wandb/run-20260814_213153-q9vcmje5/files/output.log

@@ -0,0 +1,112 @@
+  0%|          | 0/1000 [00:00<?, ?it/s][transformers] `use_return_dict` is deprecated! Use `return_dict` instead!
+[INFO] Causal mask (float with -inf) applied to all attention layers.
+ 10%|█         | 100/1000 [01:46<15:55,  1.06s/it][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': '28.88', 'grad_norm': '4', 'learning_rate': '0.001', 'epoch': '0.01078', 'train/total_time_seconds': '13.99', 'train/time_per_step_avg': '0.6996', 'train/epoch_time_elapsed': '21.69', 'train/estimated_remaining_minutes': '11.43'}
+{'loss': '24.01', 'grad_norm': '1.398', 'learning_rate': '0.001', 'epoch': '0.02157', 'train/total_time_seconds': '27.59', 'train/time_per_step_avg': '0.6897', 'train/epoch_time_elapsed': '42.88', 'train/estimated_remaining_minutes': '11.04'}
+{'loss': '22.36', 'grad_norm': '1.453', 'learning_rate': '0.001', 'epoch': '0.03235', 'train/total_time_seconds': '41.22', 'train/time_per_step_avg': '0.6871', 'train/epoch_time_elapsed': '64.16', 'train/estimated_remaining_minutes': '10.76'}
+{'loss': '20.6', 'grad_norm': '3.344', 'learning_rate': '0.001', 'epoch': '0.04314', 'train/total_time_seconds': '54.91', 'train/time_per_step_avg': '0.6864', 'train/epoch_time_elapsed': '85.37', 'train/estimated_remaining_minutes': '10.52'}
+{'loss': '18.65', 'grad_norm': '1.914', 'learning_rate': '0.001', 'epoch': '0.05392', 'train/total_time_seconds': '68.49', 'train/time_per_step_avg': '0.6849', 'train/epoch_time_elapsed': '106.5', 'train/estimated_remaining_minutes': '10.27'}
+  - 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, 30.62it/s]
+ 20%|██        | 200/1000 [03:25<10:32,  1.26it/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': '17.36', 'grad_norm': '3.109', 'learning_rate': '0.001', 'epoch': '0.06471', 'train/total_time_seconds': '82.16', 'train/time_per_step_avg': '0.6817', 'train/epoch_time_elapsed': '127.8', 'train/estimated_remaining_minutes': '10.04'}
+{'loss': '16.4', 'grad_norm': '1.602', 'learning_rate': '0.001', 'epoch': '0.07549', 'train/total_time_seconds': '95.81', 'train/time_per_step_avg': '0.6822', 'train/epoch_time_elapsed': '149.1', 'train/estimated_remaining_minutes': '9.809'}
+{'loss': '15.57', 'grad_norm': '1.523', 'learning_rate': '0.001', 'epoch': '0.08628', 'train/total_time_seconds': '108.4', 'train/time_per_step_avg': '0.6715', 'train/epoch_time_elapsed': '169.2', 'train/estimated_remaining_minutes': '9.483'}
+{'loss': '14.91', 'grad_norm': '1.703', 'learning_rate': '0.001', 'epoch': '0.09706', 'train/total_time_seconds': '121', 'train/time_per_step_avg': '0.6604', 'train/epoch_time_elapsed': '189.3', 'train/estimated_remaining_minutes': '9.183'}
+{'loss': '14.34', 'grad_norm': '2.438', 'learning_rate': '0.001', 'epoch': '0.1078', 'train/total_time_seconds': '129', 'train/time_per_step_avg': '0.6053', 'train/epoch_time_elapsed': '205.1', 'train/estimated_remaining_minutes': '8.601'}
+  - 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, 80.79it/s]
+ 30%|███       | 300/1000 [04:41<08:55,  1.31it/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': '13.84', 'grad_norm': '3.062', 'learning_rate': '0.001', 'epoch': '0.1186', 'train/total_time_seconds': '136.7', 'train/time_per_step_avg': '0.5454', 'train/epoch_time_elapsed': '220.5', 'train/estimated_remaining_minutes': '8.078'}
+{'loss': '13.42', 'grad_norm': '1.57', 'learning_rate': '0.001', 'epoch': '0.1294', 'train/total_time_seconds': '144.4', 'train/time_per_step_avg': '0.4856', 'train/epoch_time_elapsed': '235.8', 'train/estimated_remaining_minutes': '7.619'}
+{'loss': '12.98', 'grad_norm': '1.969', 'learning_rate': '0.001', 'epoch': '0.1402', 'train/total_time_seconds': '152', 'train/time_per_step_avg': '0.4363', 'train/epoch_time_elapsed': '251.1', 'train/estimated_remaining_minutes': '7.211'}
+{'loss': '12.63', 'grad_norm': '2.578', 'learning_rate': '0.001', 'epoch': '0.151', 'train/total_time_seconds': '159.7', 'train/time_per_step_avg': '0.3872', 'train/epoch_time_elapsed': '266.5', 'train/estimated_remaining_minutes': '6.843'}
+{'loss': '12.3', 'grad_norm': '2.078', 'learning_rate': '0.001', 'epoch': '0.1618', 'train/total_time_seconds': '167.3', 'train/time_per_step_avg': '0.3829', 'train/epoch_time_elapsed': '281.7', 'train/estimated_remaining_minutes': '6.507'}
+  - 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, 78.16it/s]
+ 40%|████      | 400/1000 [05:57<07:30,  1.33it/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': '11.98', 'grad_norm': '2.094', 'learning_rate': '0.001', 'epoch': '0.1726', 'train/total_time_seconds': '175', 'train/time_per_step_avg': '0.3829', 'train/epoch_time_elapsed': '296.9', 'train/estimated_remaining_minutes': '6.197'}
+{'loss': '11.69', 'grad_norm': '2.172', 'learning_rate': '0.001', 'epoch': '0.1833', 'train/total_time_seconds': '182.7', 'train/time_per_step_avg': '0.383', 'train/epoch_time_elapsed': '312.2', 'train/estimated_remaining_minutes': '5.91'}
+{'loss': '11.42', 'grad_norm': '2.125', 'learning_rate': '0.001', 'epoch': '0.1941', 'train/total_time_seconds': '190.3', 'train/time_per_step_avg': '0.3832', 'train/epoch_time_elapsed': '327.6', 'train/estimated_remaining_minutes': '5.639'}
+{'loss': '11.16', 'grad_norm': '2.234', 'learning_rate': '0.001', 'epoch': '0.2049', 'train/total_time_seconds': '198', 'train/time_per_step_avg': '0.3831', 'train/epoch_time_elapsed': '342.7', 'train/estimated_remaining_minutes': '5.384'}
+{'loss': '10.94', 'grad_norm': '2.453', 'learning_rate': '0.001', 'epoch': '0.2157', 'train/total_time_seconds': '205.6', 'train/time_per_step_avg': '0.3832', 'train/epoch_time_elapsed': '357.8', 'train/estimated_remaining_minutes': '5.141'}
+  - 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, 77.92it/s]
+ 50%|█████     | 500/1000 [07:14<06:18,  1.32it/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': '10.73', 'grad_norm': '2.078', 'learning_rate': '0.001', 'epoch': '0.2265', 'train/total_time_seconds': '213.3', 'train/time_per_step_avg': '0.3829', 'train/epoch_time_elapsed': '373.1', 'train/estimated_remaining_minutes': '4.909'}
+{'loss': '10.6', 'grad_norm': '2.719', 'learning_rate': '0.001', 'epoch': '0.2373', 'train/total_time_seconds': '220.9', 'train/time_per_step_avg': '0.3828', 'train/epoch_time_elapsed': '388.4', 'train/estimated_remaining_minutes': '4.687'}
+{'loss': '10.4', 'grad_norm': '2.516', 'learning_rate': '0.001', 'epoch': '0.248', 'train/total_time_seconds': '228.6', 'train/time_per_step_avg': '0.3826', 'train/epoch_time_elapsed': '403.6', 'train/estimated_remaining_minutes': '4.472'}
+{'loss': '10.24', 'grad_norm': '2.391', 'learning_rate': '0.001', 'epoch': '0.2588', 'train/total_time_seconds': '236.2', 'train/time_per_step_avg': '0.3826', 'train/epoch_time_elapsed': '418.8', 'train/estimated_remaining_minutes': '4.265'}
+{'loss': '10.09', 'grad_norm': '2.25', 'learning_rate': '0.001', 'epoch': '0.2696', 'train/total_time_seconds': '243.9', 'train/time_per_step_avg': '0.3826', 'train/epoch_time_elapsed': '434', 'train/estimated_remaining_minutes': '4.065'}
+  - 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, 58.13it/s]
+ 60%|██████    | 600/1000 [08:30<05:05,  1.31it/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': '9.968', 'grad_norm': '2.219', 'learning_rate': '0.001', 'epoch': '0.2804', 'train/total_time_seconds': '251.5', 'train/time_per_step_avg': '0.3826', 'train/epoch_time_elapsed': '449.3', 'train/estimated_remaining_minutes': '3.87'}
+{'loss': '9.856', 'grad_norm': '2.016', 'learning_rate': '0.001', 'epoch': '0.2912', 'train/total_time_seconds': '259.2', 'train/time_per_step_avg': '0.3825', 'train/epoch_time_elapsed': '464.5', 'train/estimated_remaining_minutes': '3.68'}
+{'loss': '9.74', 'grad_norm': '2.438', 'learning_rate': '0.001', 'epoch': '0.302', 'train/total_time_seconds': '266.8', 'train/time_per_step_avg': '0.3826', 'train/epoch_time_elapsed': '479.6', 'train/estimated_remaining_minutes': '3.494'}
+{'loss': '9.627', 'grad_norm': '1.977', 'learning_rate': '0.001', 'epoch': '0.3128', 'train/total_time_seconds': '274.5', 'train/time_per_step_avg': '0.3826', 'train/epoch_time_elapsed': '495', 'train/estimated_remaining_minutes': '3.313'}
+{'loss': '9.515', 'grad_norm': '2.109', 'learning_rate': '0.001', 'epoch': '0.3235', 'train/total_time_seconds': '282.2', 'train/time_per_step_avg': '0.3827', 'train/epoch_time_elapsed': '510.2', 'train/estimated_remaining_minutes': '3.135'}
+  - 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, 61.34it/s]
+ 70%|███████   | 700/1000 [09:46<03:53,  1.29it/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': '9.426', 'grad_norm': '2.203', 'learning_rate': '0.001', 'epoch': '0.3343', 'train/total_time_seconds': '289.8', 'train/time_per_step_avg': '0.3831', 'train/epoch_time_elapsed': '525.7', 'train/estimated_remaining_minutes': '2.961'}
+{'loss': '9.341', 'grad_norm': '2.297', 'learning_rate': '0.001', 'epoch': '0.3451', 'train/total_time_seconds': '297.5', 'train/time_per_step_avg': '0.3832', 'train/epoch_time_elapsed': '540.9', 'train/estimated_remaining_minutes': '2.789'}
+{'loss': '9.255', 'grad_norm': '1.82', 'learning_rate': '0.001', 'epoch': '0.3559', 'train/total_time_seconds': '305.2', 'train/time_per_step_avg': '0.3833', 'train/epoch_time_elapsed': '556.2', 'train/estimated_remaining_minutes': '2.62'}
+{'loss': '9.18', 'grad_norm': '2.484', 'learning_rate': '0.001', 'epoch': '0.3667', 'train/total_time_seconds': '312.8', 'train/time_per_step_avg': '0.3832', 'train/epoch_time_elapsed': '571.3', 'train/estimated_remaining_minutes': '2.454'}
+{'loss': '9.107', 'grad_norm': '2.516', 'learning_rate': '0.001', 'epoch': '0.3775', 'train/total_time_seconds': '320.5', 'train/time_per_step_avg': '0.3831', 'train/epoch_time_elapsed': '586.7', 'train/estimated_remaining_minutes': '2.289'}
+  - 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, 82.20it/s]
+ 80%|████████  | 800/1000 [11:02<02:30,  1.33it/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': '9.01', 'grad_norm': '2', 'learning_rate': '0.001', 'epoch': '0.3882', 'train/total_time_seconds': '328.1', 'train/time_per_step_avg': '0.3826', 'train/epoch_time_elapsed': '601.9', 'train/estimated_remaining_minutes': '2.127'}
+{'loss': '8.977', 'grad_norm': '2.156', 'learning_rate': '0.001', 'epoch': '0.399', 'train/total_time_seconds': '335.8', 'train/time_per_step_avg': '0.3825', 'train/epoch_time_elapsed': '617.3', 'train/estimated_remaining_minutes': '1.966'}
+{'loss': '8.914', 'grad_norm': '1.961', 'learning_rate': '0.001', 'epoch': '0.4098', 'train/total_time_seconds': '343.4', 'train/time_per_step_avg': '0.3821', 'train/epoch_time_elapsed': '632.4', 'train/estimated_remaining_minutes': '1.807'}
+{'loss': '8.872', 'grad_norm': '1.883', 'learning_rate': '0.001', 'epoch': '0.4206', 'train/total_time_seconds': '351', 'train/time_per_step_avg': '0.382', 'train/epoch_time_elapsed': '647.6', 'train/estimated_remaining_minutes': '1.65'}
+{'loss': '8.77', 'grad_norm': '2', 'learning_rate': '0.001', 'epoch': '0.4314', 'train/total_time_seconds': '358.7', 'train/time_per_step_avg': '0.3819', 'train/epoch_time_elapsed': '662.7', 'train/estimated_remaining_minutes': '1.494'}
+  - 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, 80.58it/s]
+ 90%|█████████ | 900/1000 [12:19<01:15,  1.32it/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': '8.745', 'grad_norm': '2', 'learning_rate': '0.001', 'epoch': '0.4422', 'train/total_time_seconds': '366.3', 'train/time_per_step_avg': '0.382', 'train/epoch_time_elapsed': '678.1', 'train/estimated_remaining_minutes': '1.34'}
+{'loss': '8.702', 'grad_norm': '2.062', 'learning_rate': '0.001', 'epoch': '0.453', 'train/total_time_seconds': '374', 'train/time_per_step_avg': '0.3822', 'train/epoch_time_elapsed': '693.3', 'train/estimated_remaining_minutes': '1.187'}
+{'loss': '8.637', 'grad_norm': '2.172', 'learning_rate': '0.001', 'epoch': '0.4637', 'train/total_time_seconds': '381.6', 'train/time_per_step_avg': '0.3824', 'train/epoch_time_elapsed': '708.7', 'train/estimated_remaining_minutes': '1.035'}
+{'loss': '8.599', 'grad_norm': '2.078', 'learning_rate': '0.001', 'epoch': '0.4745', 'train/total_time_seconds': '389.3', 'train/time_per_step_avg': '0.3823', 'train/epoch_time_elapsed': '723.8', 'train/estimated_remaining_minutes': '0.8847'}
+{'loss': '8.55', 'grad_norm': '2.062', 'learning_rate': '0.001', 'epoch': '0.4853', 'train/total_time_seconds': '396.9', 'train/time_per_step_avg': '0.3822', 'train/epoch_time_elapsed': '739', 'train/estimated_remaining_minutes': '0.735'}
+  - 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, 48.84it/s]
+100%|██████████| 1000/1000 [13:44<00:00,  1.33i[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': '8.507', 'grad_norm': '2.016', 'learning_rate': '0.001', 'epoch': '0.4961', 'train/total_time_seconds': '404.5', 'train/time_per_step_avg': '0.3821', 'train/epoch_time_elapsed': '754.2', 'train/estimated_remaining_minutes': '0.5863'}
+{'loss': '8.488', 'grad_norm': '1.906', 'learning_rate': '0.001', 'epoch': '0.5069', 'train/total_time_seconds': '412.2', 'train/time_per_step_avg': '0.3818', 'train/epoch_time_elapsed': '769.5', 'train/estimated_remaining_minutes': '0.4385'}
+{'loss': '8.468', 'grad_norm': '1.867', 'learning_rate': '0.001', 'epoch': '0.5177', 'train/total_time_seconds': '419.8', 'train/time_per_step_avg': '0.3816', 'train/epoch_time_elapsed': '784.5', 'train/estimated_remaining_minutes': '0.2915'}
+{'loss': '8.435', 'grad_norm': '1.672', 'learning_rate': '0.001', 'epoch': '0.5284', 'train/total_time_seconds': '427.4', 'train/time_per_step_avg': '0.3817', 'train/epoch_time_elapsed': '799.7', 'train/estimated_remaining_minutes': '0.1454'}
+{'loss': '8.396', 'grad_norm': '2.094', 'learning_rate': '0.001', 'epoch': '0.5392', 'train/total_time_seconds': '435.1', 'train/time_per_step_avg': '0.3818', 'train/epoch_time_elapsed': '814.8', 'train/estimated_remaining_minutes': '0'}
+  - 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': '2.101', 'eval_runtime': '9.424', 'eval_samples_per_second': '1011', 'eval_steps_per_second': '7.958', 'epoch': '0.5392', 'train/total_time_seconds': '435.1', 'train/time_per_step_avg': '0.3818', 'train/epoch_time_elapsed': '824.2', 'train/estimated_remaining_minutes': '0'}
+  - 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, 84.50it/s]
+100%|██████████| 1000/1000 [13:44<00:00,  1.21it/s], ?it/s]
+{'train_runtime': '825.2', 'train_samples_per_second': '620.5', 'train_steps_per_second': '1.212', 'train_loss': '11.85', 'epoch': '0.5392', 'train/total_time_seconds': '435.1', 'train/time_per_step_avg': '0.3818', 'train/epoch_time_elapsed': '824.3', 'train/estimated_remaining_minutes': '0'}
+100%|██████████| 75/75 [00:09<00:00,  7.90it/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.
+  - 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, 73.91it/s]
+>>> FINISHED glu-waleed10-21L successfully

zain/Activation/wandb/run-20260814_213153-q9vcmje5/files/requirements.txt

@@ -0,0 +1,149 @@
+asttokens==3.0.1
+comm==0.2.3
+debugpy==1.8.21
+decorator==5.3.1
+executing==2.2.1
+nest-asyncio==1.6.0
+parso==0.8.7
+platformdirs==4.11.0
+psutil==7.2.2
+ptyprocess==0.7.0
+pure_eval==0.2.3
+Pygments==2.20.0
+pyzmq==27.1.0
+setuptools==83.0.0
+six==1.17.0
+tornado==6.5.7
+traitlets==5.15.0
+fsspec==2026.4.0
+wcwidth==0.8.2
+ipython_pygments_lexers==1.1.1
+jedi==0.20.0
+jupyter_core==5.9.1
+matplotlib-inline==0.2.2
+pexpect==4.9.0
+prompt_toolkit==3.0.53
+python-dateutil==2.9.0.post0
+stack_data==0.6.3
+wheel==0.47.0
+jupyter_client==8.9.1
+pip==26.1.2
+ipython==9.15.0
+ipykernel==7.2.0
+threadpoolctl==3.6.0
+pyparsing==3.3.2
+typing_extensions==4.15.0
+Jinja2==3.1.6
+narwhals==2.24.0
+kiwisolver==1.5.0
+joblib==1.5.3
+fonttools==4.63.0
+cycler==0.12.1
+scipy==1.17.1
+pandas==3.0.5
+contourpy==1.3.3
+scikit-learn==1.9.0
+matplotlib==3.11.1
+urllib3==2.7.0
+tqdm==4.70.0
+idna==3.18
+charset-normalizer==3.4.9
+certifi==2026.7.22
+requests==2.34.2
+seaborn==0.13.2
+uv==0.12.0
+shellingham==1.5.4
+mpmath==1.3.0
+attrs==26.1.0
+hf-xet==1.5.2
+nvidia-nccl-cu12==2.21.5
+MarkupSafe==3.0.3
+regex==2026.7.19
+importlib_metadata==9.0.0
+httpcore==1.0.9
+annotated-doc==0.0.5
+multidict==6.7.1
+aiohttp==3.14.3
+aiosignal==1.4.0
+xxhash==3.8.1
+aiohappyeyeballs==2.7.1
+mdurl==0.1.2
+cuda-toolkit==13.0.3.0
+networkx==3.6.1
+PyYAML==6.0.3
+nvidia-cufile==1.15.1.6
+typer==0.27.0
+torchaudio==2.6.0+cu124
+rich==15.0.0
+nvidia-cufft-cu12==11.2.1.3
+h11==0.16.0
+dill==0.4.1
+cuda-pathfinder==1.6.0
+filelock==3.29.0
+nvidia-nvtx-cu12==12.4.127
+httpx==0.28.1
+anyio==4.14.2
+numpy==2.4.4
+yarl==1.24.5
+click==8.4.2
+triton==3.2.0
+frozenlist==1.8.0
+zipp==4.1.0
+propcache==0.5.2
+tokenizers==0.22.2
+markdown-it-py==4.2.0
+nvidia-cuda-runtime==13.0.96
+cuda-bindings==13.3.1
+nvidia-cuda-cupti==13.0.85
+torch==2.6.0+cu124
+multiprocess==0.70.19
+pillow==12.2.0
+transformers==5.16.0.dev0
+wandb==0.28.1
+nvidia-curand==10.4.0.35
+sympy==1.13.1
+nvidia-cusparse==12.6.3.3
+nvidia-cuda-nvrtc==13.0.88
+typing-inspection==0.4.2
+nvidia-cusolver==12.0.4.66
+nvidia-cufft==12.0.0.61
+nvidia-cudnn-cu13==9.20.0.48
+nvidia-cublas==13.1.1.3
+pyarrow==25.0.0
+evaluate==0.4.6
+diffusers==0.39.0
+pydantic==2.13.4
+annotated-types==0.8.0
+protobuf==7.35.1
+sentry-sdk==2.66.1
+einops==0.8.2
+packaging==26.2
+nvidia-nvjitlink-cu12==12.4.127
+nvidia-curand-cu12==10.3.5.147
+nvidia-cusparselt-cu12==0.6.2
+nvidia-cusparse-cu12==12.3.1.170
+nvidia-cuda-runtime-cu12==12.4.127
+torchvision==0.21.0+cu124
+nvidia-cuda-nvrtc-cu12==12.4.127
+nvidia-cuda-cupti-cu12==12.4.127
+nvidia-cusolver-cu12==11.6.1.9
+nvidia-cublas-cu12==12.4.5.8
+nvidia-cudnn-cu12==9.1.0.70
+huggingface_hub==1.26.0
+datasets==5.0.1
+safetensors==0.8.0
+accelerate==1.14.0
+pydantic_core==2.46.4
+ninja==1.13.0
+autocommand==2.2.2
+backports.tarfile==1.2.0
+importlib_metadata==8.7.1
+jaraco.text==4.0.0
+jaraco.context==6.1.0
+jaraco.functools==4.4.0
+more-itertools==10.8.0
+packaging==26.0
+platformdirs==4.4.0
+tomli==2.4.0
+wheel==0.46.3
+zipp==3.23.0

zain/Activation/wandb/run-20260814_213153-q9vcmje5/files/wandb-metadata.json

@@ -0,0 +1,101 @@
+{
+  "os":  "Linux-5.15.0-126-generic-x86_64-with-glibc2.35",
+  "python":  "CPython 3.11.15",
+  "startedAt":  "2026-08-14T21:31:53.799629Z",
+  "args":  [
+    "--config",
+    "/mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/configs/baseline.yaml",
+    "--variants",
+    "glu-waleed10",
+    "glu-silu-waleed10",
+    "mlp-waleed10",
+    "mlp-silu-waleed10",
+    "glu-waleed",
+    "glu-situglu_low",
+    "glu-waleedglu_low"
+  ],
+  "program":  "/mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/sweep.py",
+  "codePath":  "sweep.py",
+  "codePathLocal":  "sweep.py",
+  "git":  {
+    "remote":  "https://github.com/w-ahmad1a10/Activation.git",
+    "commit":  "3fb47c6943d6927eafc39bd399a3d3a520bae74a"
+  },
+  "email":  "deepnevro@gmail.com",
+  "root":  "/mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation",
+  "host":  "deeplens-k3s-node1",
+  "executable":  "/mnt/data/zainulabideen/zain-exp/notebooks/my_env/bin/python",
+  "cpu_count":  112,
+  "cpu_count_logical":  224,
+  "gpu":  "NVIDIA H100 80GB HBM3",
+  "gpu_count":  8,
+  "disk":  {
+    "/":  {
+      "total":  "1560765693952",
+      "used":  "709060804608"
+    }
+  },
+  "memory":  {
+    "total":  "2164089937920"
+  },
+  "gpu_nvidia":  [
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-39c684a5-fde6-83d7-1663-0859795881ae"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-68012e5a-38b6-b643-0ca6-62fb66720bf3"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-132944c4-b689-2b5f-89a4-d730401677ab"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-2df386cc-6d26-d0e2-7a2d-a057b0d95864"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-bfa16575-1d94-1aa2-4537-2c93433f42ef"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-bc6c3e3c-9b90-09ca-c034-774961847c54"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-00a441e1-7c95-e7d6-4c35-43d6b291aea9"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-1c4d29a2-4647-6fce-d8fc-0c5ecfbbd6ea"
+    }
+  ],
+  "cudaVersion":  "12.4",
+  "writerId":  "avd7wjdsk6ji6qbb7ojy8wwedk5izmcm"
+}

zain/Activation/wandb/run-20260814_213153-q9vcmje5/files/wandb-summary.json

@@ -0,0 +1 @@
+{"train/global_step":1000,"train_runtime":825.1777,"_runtime":834,"train/epoch":0.5392289026691831,"train_samples_per_second":620.472,"train_loss":11.851402679443359,"train/grad_norm":2.09375,"train_steps_per_second":1.212,"eval/loss":2.1011769771575928,"total_flos":5.420315836416e+15,"eval/samples_per_second":982.674,"eval/steps_per_second":7.736,"eval/runtime":9.695,"_step":52,"train/learning_rate":0.001,"_wandb":{"runtime":834},"train/loss":8.39569091796875,"_timestamp":1.7867439483697834e+09}

zain/Activation/wandb/run-20260814_213153-q9vcmje5/logs/debug-core.log

@@ -0,0 +1,62 @@
+{"time":"2026-08-14T21:31:53.539152008Z","level":"INFO","msg":"main: starting server","port-filename":"/tmp/tmp650y3sc2/port-1229423.txt","pid":1229423,"detached":false,"idle-timeout":600000000000,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false,"enable-dcgm-profiling":false}
+{"time":"2026-08-14T21:31:53.540676876Z","level":"INFO","msg":"server: will exit if parent process dies","ppid":1229423}
+{"time":"2026-08-14T21:31:53.540659035Z","level":"INFO","msg":"server: accepting connections","addr":{"Name":"/tmp/wandb-1229423-1230766-2604858008/socket","Net":"unix"}}
+{"time":"2026-08-14T21:31:53.717724383Z","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"1(@)"}
+{"time":"2026-08-14T21:31:53.802092564Z","level":"INFO","msg":"handleInformInit: received","streamId":"q9vcmje5","id":"1(@)"}
+{"time":"2026-08-14T21:31:54.062397411Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"q9vcmje5","id":"1(@)"}
+{"time":"2026-08-14T21:31:59.411924652Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"32s1adqk73rr"}
+{"time":"2026-08-14T21:45:48.396635188Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"32s1adqk73rr"}
+{"time":"2026-08-14T21:45:48.929834053Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"32s1adqk73rr"}
+{"time":"2026-08-14T21:45:48.931620539Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"q9vcmje5","id":"1(@)"}
+{"time":"2026-08-14T21:45:48.932491197Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"q9vcmje5","id":"1(@)"}
+{"time":"2026-08-14T21:45:49.464388506Z","level":"INFO","msg":"handleInformInit: received","streamId":"k24y90ws","id":"1(@)"}
+{"time":"2026-08-14T21:45:49.733853364Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"k24y90ws","id":"1(@)"}
+{"time":"2026-08-14T21:45:55.329792921Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"dr322rnt7xgx"}
+{"time":"2026-08-14T21:59:01.158537572Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"dr322rnt7xgx"}
+{"time":"2026-08-14T21:59:01.914569559Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"dr322rnt7xgx"}
+{"time":"2026-08-14T21:59:01.916156477Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"k24y90ws","id":"1(@)"}
+{"time":"2026-08-14T21:59:01.917534874Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"k24y90ws","id":"1(@)"}
+{"time":"2026-08-14T21:59:02.524868682Z","level":"INFO","msg":"handleInformInit: received","streamId":"npap5k3g","id":"1(@)"}
+{"time":"2026-08-14T21:59:02.786199875Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"npap5k3g","id":"1(@)"}
+{"time":"2026-08-14T21:59:08.192858634Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"2g99yy5pussw"}
+{"time":"2026-08-14T22:11:53.643379161Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"2g99yy5pussw"}
+{"time":"2026-08-14T22:11:54.173579992Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"2g99yy5pussw"}
+{"time":"2026-08-14T22:11:54.17490129Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"npap5k3g","id":"1(@)"}
+{"time":"2026-08-14T22:11:54.175479186Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"npap5k3g","id":"1(@)"}
+{"time":"2026-08-14T22:11:54.694802218Z","level":"INFO","msg":"handleInformInit: received","streamId":"a3607q94","id":"1(@)"}
+{"time":"2026-08-14T22:11:54.956642682Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"a3607q94","id":"1(@)"}
+{"time":"2026-08-14T22:12:00.441973218Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"raaqqqmajsfs"}
+{"time":"2026-08-14T22:24:55.662446346Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"raaqqqmajsfs"}
+{"time":"2026-08-14T22:24:56.296327526Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"raaqqqmajsfs"}
+{"time":"2026-08-14T22:24:56.297810539Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"a3607q94","id":"1(@)"}
+{"time":"2026-08-14T22:24:56.29856977Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"a3607q94","id":"1(@)"}
+{"time":"2026-08-14T22:24:56.82560196Z","level":"INFO","msg":"handleInformInit: received","streamId":"wcvy5vyw","id":"1(@)"}
+{"time":"2026-08-14T22:24:57.085772231Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"wcvy5vyw","id":"1(@)"}
+{"time":"2026-08-14T22:25:02.571913681Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"2nb8wusqu4fo"}
+{"time":"2026-08-14T22:38:28.612623637Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"2nb8wusqu4fo"}
+{"time":"2026-08-14T22:38:29.208727987Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"2nb8wusqu4fo"}
+{"time":"2026-08-14T22:38:29.210106262Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"wcvy5vyw","id":"1(@)"}
+{"time":"2026-08-14T22:38:29.210705504Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"wcvy5vyw","id":"1(@)"}
+{"time":"2026-08-14T22:38:29.839594174Z","level":"INFO","msg":"handleInformInit: received","streamId":"uuv4h936","id":"1(@)"}
+{"time":"2026-08-14T22:38:30.099444184Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"uuv4h936","id":"1(@)"}
+{"time":"2026-08-14T22:38:35.620666929Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"r0aouorh27ls"}
+{"time":"2026-08-14T22:52:17.704425638Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"r0aouorh27ls"}
+{"time":"2026-08-14T22:52:18.35448297Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"r0aouorh27ls"}
+{"time":"2026-08-14T22:52:18.355720686Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"uuv4h936","id":"1(@)"}
+{"time":"2026-08-14T22:52:18.356578583Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"uuv4h936","id":"1(@)"}
+{"time":"2026-08-14T22:52:18.899799881Z","level":"INFO","msg":"handleInformInit: received","streamId":"x87hwrtd","id":"1(@)"}
+{"time":"2026-08-14T22:52:19.159494071Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"x87hwrtd","id":"1(@)"}
+{"time":"2026-08-14T22:52:24.671083701Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"hn7lqjhi3ekb"}
+{"time":"2026-08-14T23:05:51.214192173Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"hn7lqjhi3ekb"}
+{"time":"2026-08-14T23:05:51.783107435Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"hn7lqjhi3ekb"}
+{"time":"2026-08-14T23:05:51.784410941Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"x87hwrtd","id":"1(@)"}
+{"time":"2026-08-14T23:05:51.78516667Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"x87hwrtd","id":"1(@)"}
+{"time":"2026-08-14T23:05:52.088192277Z","level":"INFO","msg":"handleInformTeardown: server teardown initiated","id":"1(@)"}
+{"time":"2026-08-14T23:05:52.088229648Z","level":"INFO","msg":"handleInformTeardown: server shutdown complete","id":"1(@)"}
+{"time":"2026-08-14T23:05:52.088239651Z","level":"INFO","msg":"connection: closing","id":"1(@)"}
+{"time":"2026-08-14T23:05:52.088243873Z","level":"INFO","msg":"server: is shutting down"}
+{"time":"2026-08-14T23:05:52.0882852Z","level":"INFO","msg":"connection: closed successfully","id":"1(@)"}
+{"time":"2026-08-14T23:05:52.088299095Z","level":"INFO","msg":"processOutgoingData: finished","id":"1(@)"}
+{"time":"2026-08-14T23:05:52.088345917Z","level":"INFO","msg":"connection: ManageConnectionData: connection closed","id":"1(@)"}
+{"time":"2026-08-14T23:05:52.088452051Z","level":"INFO","msg":"server: listener closed","addr":{"Name":"/tmp/wandb-1229423-1230766-2604858008/socket","Net":"unix"}}
+{"time":"2026-08-14T23:05:52.088520571Z","level":"INFO","msg":"server: all connections closed"}

zain/Activation/wandb/run-20260814_213153-q9vcmje5/logs/debug-internal.log

@@ -0,0 +1,127 @@
+{"time":"2026-08-14T21:31:53.802248642Z","level":"INFO","msg":"wandb-core"}
+{"time":"2026-08-14T21:31:53.802536305Z","level":"INFO","msg":"stream: starting","core version":"0.28.1"}
+{"time":"2026-08-14T21:31:54.062231678Z","level":"INFO","msg":"stream: created new stream","id":"q9vcmje5"}
+{"time":"2026-08-14T21:31:54.062319994Z","level":"INFO","msg":"handler: started"}
+{"time":"2026-08-14T21:31:54.062390309Z","level":"INFO","msg":"stream: started"}
+{"time":"2026-08-14T21:31:54.062412718Z","level":"INFO","msg":"writer: started","stream_id":"q9vcmje5"}
+{"time":"2026-08-14T21:31:54.062420181Z","level":"INFO","msg":"sender: started"}
+{"time":"2026-08-14T21:31:54.416838412Z","level":"INFO","msg":"filestream: sending request","total_files":1,"console_offset":0,"console_lines":1}
+{"time":"2026-08-14T21:31:54.510348438Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:32:09.41771802Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":0,"events_lines":1,"console_offset":0,"console_lines":3,"uploaded_len":2}
+{"time":"2026-08-14T21:32:09.578280298Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:32:24.417605975Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":0,"history_lines":1,"events_offset":1,"events_lines":2,"console_offset":2,"console_lines":2}
+{"time":"2026-08-14T21:32:24.549935438Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:32:39.41719538Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":1,"history_lines":1,"events_offset":3,"events_lines":2,"console_offset":2,"console_lines":1}
+{"time":"2026-08-14T21:32:39.566802488Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:32:54.417058222Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":5,"events_lines":2,"console_offset":2,"console_lines":1}
+{"time":"2026-08-14T21:32:54.526885785Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:33:09.417361811Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":2,"history_lines":1,"events_offset":7,"events_lines":2,"console_offset":2,"console_lines":1}
+{"time":"2026-08-14T21:33:09.541292122Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:33:24.41766197Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":3,"history_lines":1,"events_offset":9,"events_lines":2,"console_offset":2,"console_lines":1}
+{"time":"2026-08-14T21:33:24.549198743Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:33:39.417110044Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":11,"events_lines":2,"console_offset":2,"console_lines":1}
+{"time":"2026-08-14T21:33:39.551776959Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:33:54.417538658Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":4,"history_lines":1,"events_offset":13,"events_lines":2,"console_offset":2,"console_lines":1}
+{"time":"2026-08-14T21:33:54.555356936Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:34:09.417890895Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":5,"history_lines":1,"events_offset":15,"events_lines":2,"console_offset":4,"console_lines":10}
+{"time":"2026-08-14T21:34:09.592684206Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:34:24.417179414Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":6,"history_lines":1,"events_offset":17,"events_lines":2,"console_offset":12,"console_lines":1}
+{"time":"2026-08-14T21:34:24.534710617Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:34:39.417073445Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":19,"events_lines":2,"console_offset":12,"console_lines":1}
+{"time":"2026-08-14T21:34:39.570434803Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:34:54.417818476Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":7,"history_lines":1,"events_offset":21,"events_lines":2,"console_offset":12,"console_lines":1}
+{"time":"2026-08-14T21:34:54.541208219Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:35:09.41711858Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":8,"history_lines":1,"events_offset":23,"events_lines":2,"console_offset":12,"console_lines":1}
+{"time":"2026-08-14T21:35:09.54503893Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:35:24.417709905Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":9,"history_lines":1,"events_offset":25,"events_lines":2,"console_offset":12,"console_lines":1}
+{"time":"2026-08-14T21:35:24.538800467Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:35:39.417929262Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":10,"history_lines":1,"events_offset":27,"events_lines":2,"console_offset":14,"console_lines":10}
+{"time":"2026-08-14T21:35:39.553593527Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:35:54.417850428Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":11,"history_lines":1,"events_offset":29,"events_lines":2,"console_offset":22,"console_lines":1}
+{"time":"2026-08-14T21:35:54.591775915Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:36:09.417299708Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":12,"history_lines":1,"events_offset":31,"events_lines":2,"console_offset":22,"console_lines":1}
+{"time":"2026-08-14T21:36:09.609766076Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:36:24.417329624Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":13,"history_lines":1,"events_offset":33,"events_lines":2,"console_offset":22,"console_lines":1}
+{"time":"2026-08-14T21:36:24.528837524Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:36:39.417160496Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":14,"history_lines":1,"events_offset":35,"events_lines":2,"console_offset":22,"console_lines":1}
+{"time":"2026-08-14T21:36:39.539248212Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:36:54.417072248Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":15,"history_lines":1,"events_offset":37,"events_lines":2,"console_offset":24,"console_lines":10}
+{"time":"2026-08-14T21:36:54.609691461Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:37:09.417049312Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":16,"history_lines":1,"events_offset":39,"events_lines":2,"console_offset":32,"console_lines":1}
+{"time":"2026-08-14T21:37:09.684843081Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:37:24.417401277Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":17,"history_lines":1,"events_offset":41,"events_lines":2,"console_offset":32,"console_lines":1}
+{"time":"2026-08-14T21:37:24.537403552Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:37:39.417411849Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":18,"history_lines":1,"events_offset":43,"events_lines":2,"console_offset":32,"console_lines":1}
+{"time":"2026-08-14T21:37:39.548230786Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:37:54.417198892Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":19,"history_lines":1,"events_offset":45,"events_lines":2,"console_offset":32,"console_lines":1}
+{"time":"2026-08-14T21:37:54.542467252Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:38:09.417264097Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":20,"history_lines":1,"events_offset":47,"events_lines":2,"console_offset":34,"console_lines":10}
+{"time":"2026-08-14T21:38:09.674000615Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:38:24.417681287Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":21,"history_lines":1,"events_offset":49,"events_lines":2,"console_offset":42,"console_lines":1}
+{"time":"2026-08-14T21:38:24.609965411Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:38:39.416972046Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":22,"history_lines":1,"events_offset":51,"events_lines":2,"console_offset":42,"console_lines":1}
+{"time":"2026-08-14T21:38:39.549505398Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:38:54.41796769Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":23,"history_lines":1,"events_offset":53,"events_lines":2,"console_offset":42,"console_lines":1}
+{"time":"2026-08-14T21:38:54.528895634Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:39:09.417164029Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":24,"history_lines":1,"events_offset":55,"events_lines":2,"console_offset":42,"console_lines":1}
+{"time":"2026-08-14T21:39:09.602030487Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:39:24.417256613Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":25,"history_lines":1,"events_offset":57,"events_lines":2,"console_offset":44,"console_lines":10}
+{"time":"2026-08-14T21:39:24.553135287Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:39:39.417138993Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":26,"history_lines":1,"events_offset":59,"events_lines":2,"console_offset":52,"console_lines":1}
+{"time":"2026-08-14T21:39:39.529219292Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:39:54.417226385Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":27,"history_lines":1,"events_offset":61,"events_lines":2,"console_offset":52,"console_lines":1}
+{"time":"2026-08-14T21:39:54.536465485Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:40:09.417043824Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":28,"history_lines":1,"events_offset":63,"events_lines":2,"console_offset":52,"console_lines":1}
+{"time":"2026-08-14T21:40:09.68305832Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:40:24.41710473Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":65,"events_lines":2,"console_offset":52,"console_lines":1}
+{"time":"2026-08-14T21:40:24.529074175Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:40:39.417102784Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":29,"history_lines":1,"events_offset":67,"events_lines":2,"console_offset":52,"console_lines":1}
+{"time":"2026-08-14T21:40:39.548617714Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:40:54.417536563Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":30,"history_lines":1,"events_offset":69,"events_lines":2,"console_offset":54,"console_lines":10}
+{"time":"2026-08-14T21:40:54.540414529Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:41:09.417723797Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":31,"history_lines":1,"events_offset":71,"events_lines":2,"console_offset":62,"console_lines":1}
+{"time":"2026-08-14T21:41:09.626110005Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:41:24.417000398Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":32,"history_lines":1,"events_offset":73,"events_lines":2,"console_offset":62,"console_lines":1}
+{"time":"2026-08-14T21:41:24.58002245Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:41:39.41726191Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":33,"history_lines":1,"events_offset":75,"events_lines":2,"console_offset":62,"console_lines":1}
+{"time":"2026-08-14T21:41:39.526092005Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:41:54.417072409Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":34,"history_lines":1,"events_offset":77,"events_lines":2,"console_offset":62,"console_lines":1}
+{"time":"2026-08-14T21:41:54.546246802Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:42:09.417123682Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":35,"history_lines":1,"events_offset":79,"events_lines":2,"console_offset":64,"console_lines":10}
+{"time":"2026-08-14T21:42:09.666318822Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:42:24.417686885Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":36,"history_lines":1,"events_offset":81,"events_lines":2,"console_offset":72,"console_lines":1}
+{"time":"2026-08-14T21:42:24.564587405Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:42:39.41722149Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":37,"history_lines":1,"events_offset":83,"events_lines":2,"console_offset":72,"console_lines":1}
+{"time":"2026-08-14T21:42:39.582243173Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:42:54.417586987Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":38,"history_lines":1,"events_offset":85,"events_lines":2,"console_offset":72,"console_lines":1}
+{"time":"2026-08-14T21:42:54.557667192Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:43:09.417207023Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":39,"history_lines":1,"events_offset":87,"events_lines":2,"console_offset":72,"console_lines":1}
+{"time":"2026-08-14T21:43:09.610669065Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:43:24.417672133Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":40,"history_lines":1,"events_offset":89,"events_lines":2,"console_offset":74,"console_lines":10}
+{"time":"2026-08-14T21:43:24.574587398Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:43:39.417277064Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":41,"history_lines":1,"events_offset":91,"events_lines":2,"console_offset":82,"console_lines":1}
+{"time":"2026-08-14T21:43:39.572880967Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:43:54.417379547Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":42,"history_lines":1,"events_offset":93,"events_lines":2,"console_offset":82,"console_lines":1}
+{"time":"2026-08-14T21:43:54.581169619Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:44:09.417088108Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":43,"history_lines":1,"events_offset":95,"events_lines":2,"console_offset":82,"console_lines":1}
+{"time":"2026-08-14T21:44:09.637501534Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:44:24.417552656Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":44,"history_lines":1,"events_offset":97,"events_lines":2,"console_offset":82,"console_lines":1}
+{"time":"2026-08-14T21:44:24.556148219Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:44:39.417501955Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":45,"history_lines":1,"events_offset":99,"events_lines":2,"console_offset":84,"console_lines":10}
+{"time":"2026-08-14T21:44:39.686483667Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:44:54.417133976Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":46,"history_lines":1,"events_offset":101,"events_lines":2,"console_offset":92,"console_lines":1}
+{"time":"2026-08-14T21:44:54.554631841Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:45:09.41749322Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":47,"history_lines":1,"events_offset":103,"events_lines":2,"console_offset":92,"console_lines":1}
+{"time":"2026-08-14T21:45:09.642217292Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:45:24.417641844Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":48,"history_lines":1,"events_offset":105,"events_lines":2,"console_offset":92,"console_lines":1}
+{"time":"2026-08-14T21:45:24.579697512Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:45:39.41776998Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":49,"history_lines":3,"events_offset":107,"events_lines":2,"console_offset":92,"console_lines":1}
+{"time":"2026-08-14T21:45:39.679514648Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:45:48.819105986Z","level":"INFO","msg":"fileTransfer: Close: file transfer manager closed"}
+{"time":"2026-08-14T21:45:48.819431393Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":52,"history_lines":1,"events_offset":109,"events_lines":1,"console_offset":94,"console_lines":18,"uploaded_len":3,"complete":true,"exit_code":0}
+{"time":"2026-08-14T21:45:48.927022721Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:45:48.928319845Z","level":"INFO","msg":"handler: operation stats","stats":{}}
+{"time":"2026-08-14T21:45:48.931654382Z","level":"INFO","msg":"stream: finishing up"}
+{"time":"2026-08-14T21:45:48.931674323Z","level":"INFO","msg":"handler: closed"}
+{"time":"2026-08-14T21:45:48.931746361Z","level":"INFO","msg":"sender: closed"}
+{"time":"2026-08-14T21:45:48.931750382Z","level":"INFO","msg":"stream: all finished"}

zain/Activation/wandb/run-20260814_213153-q9vcmje5/logs/debug.log

@@ -0,0 +1,27 @@
+2026-08-14 21:31:53,800 INFO    MainThread:1229423 [wandb_setup.py:_flush():81] Current SDK version is 0.28.1
+2026-08-14 21:31:53,800 INFO    MainThread:1229423 [wandb_setup.py:_flush():81] Configure stats pid to 1229423
+2026-08-14 21:31:53,800 INFO    MainThread:1229423 [wandb_setup.py:_flush():81] Loading settings from environment variables
+2026-08-14 21:31:53,800 INFO    MainThread:1229423 [wandb_init.py:setup_run_log_directory():729] Logging user logs to /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/wandb/run-20260814_213153-q9vcmje5/logs/debug.log
+2026-08-14 21:31:53,800 INFO    MainThread:1229423 [wandb_init.py:setup_run_log_directory():730] Logging internal logs to /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/wandb/run-20260814_213153-q9vcmje5/logs/debug-internal.log
+2026-08-14 21:31:53,800 INFO    MainThread:1229423 [wandb_init.py:init():772] calling init triggers
+2026-08-14 21:31:53,801 INFO    MainThread:1229423 [wandb_init.py:init():777] wandb.init called with sweep_config: {}
+config: {'_wandb': {}}
+2026-08-14 21:31:53,801 INFO    MainThread:1229423 [wandb_init.py:init():820] starting backend
+2026-08-14 21:31:53,801 INFO    MainThread:1229423 [wandb_init.py:init():835] sending inform_init request
+2026-08-14 21:31:54,062 INFO    MainThread:1229423 [wandb_init.py:init():840] backend started and connected
+2026-08-14 21:31:54,063 INFO    MainThread:1229423 [wandb_init.py:init():910] updated telemetry
+2026-08-14 21:31:54,070 INFO    MainThread:1229423 [wandb_init.py:init():933] communicating run to backend with 90.0 second timeout
+2026-08-14 21:31:54,336 INFO    MainThread:1229423 [wandb_init.py:init():978] starting run threads in backend
+2026-08-14 21:31:54,409 INFO    MainThread:1229423 [wandb_run.py:_console_start():2621] atexit reg
+2026-08-14 21:31:54,409 INFO    MainThread:1229423 [wandb_run.py:_redirect():2471] redirect: wrap_raw
+2026-08-14 21:31:54,409 INFO    MainThread:1229423 [wandb_run.py:_redirect():2540] Wrapping output streams.
+2026-08-14 21:31:54,409 INFO    MainThread:1229423 [wandb_run.py:_redirect():2563] Redirects installed.
+2026-08-14 21:31:54,412 INFO    MainThread:1229423 [wandb_init.py:init():1016] run started, returning control to user process
+2026-08-14 21:31:54,413 INFO    MainThread:1229423 [wandb_run.py:_config_callback():1346] config_cb None None {'transformers_version': '5.16.0.dev0', 'architectures': None, 'output_hidden_states': False, 'return_dict': True, 'dtype': None, 'chunk_size_feed_forward': 0, 'is_encoder_decoder': False, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'problem_type': None, 'vocab_size': 4096, 'hidden_size': 128, 'intermediate_size': 256, 'num_hidden_layers': 21, 'num_attention_heads': 4, 'num_key_value_heads': 4, 'hidden_act': 'silu', 'max_position_embeddings': 512, 'initializer_range': 0.02, 'rms_norm_eps': 1e-06, 'use_cache': False, 'pad_token_id': 0, 'bos_token_id': 1, 'eos_token_id': 2, 'pretraining_tp': 1, 'tie_word_embeddings': True, 'rope_parameters': {'rope_theta': 10000.0, 'rope_type': 'default'}, 'attention_bias': False, 'attention_dropout': 0.0, 'mlp_bias': False, 'head_dim': 32, '_name_or_path': '', 'tokenizer_name': 'w-ahmad/tiny-stories-tokenizer', 'mlp_type': 'glu', 'activation': 'waleed10', 'waleed_beta': 10.0, 'model_type': 'tiny_llama', 'output_attentions': False, 'output_dir': 'out/glu-waleed10-21L_run', 'per_device_train_batch_size': 128, 'num_train_epochs': 1, 'max_steps': 1000, 'learning_rate': 0.001, 'lr_scheduler_type': 'constant', 'lr_scheduler_kwargs': None, 'warmup_steps': 0, 'optim': 'adamw_torch', 'optim_args': None, 'weight_decay': 0.01, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'optim_target_modules': None, 'gradient_accumulation_steps': 4, 'average_tokens_across_devices': True, 'max_grad_norm': 1.0, 'label_smoothing_factor': 0.0, 'bf16': True, 'fp16': False, 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'gradient_checkpointing': False, 'gradient_checkpointing_kwargs': None, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'use_liger_kernel': False, 'liger_kernel_config': None, 'neftune_noise_alpha': None, 'torch_empty_cache_steps': None, 'auto_find_batch_size': False, 'logging_strategy': 'steps', 'logging_steps': 20, 'logging_first_step': False, 'log_on_each_node': True, 'logging_nan_inf_filter': True, 'include_num_input_tokens_seen': 'no', 'log_level': 'passive', 'log_level_replica': 'warning', 'disable_tqdm': False, 'report_to': ['wandb'], 'run_name': 'LM-glu-waleed10-21L-4.0M-20260814-213152', 'project': 'huggingface', 'trackio_space_id': None, 'trackio_bucket_id': None, 'trackio_static_space_id': None, 'eval_strategy': 'steps', 'eval_steps': 60000, 'eval_delay': 0, 'per_device_eval_batch_size': 128, 'prediction_loss_only': False, 'eval_on_start': False, 'eval_do_concat_batches': True, 'eval_use_gather_object': False, 'eval_accumulation_steps': None, 'include_for_metrics': [], 'batch_eval_metrics': False, 'save_only_model': False, 'save_strategy': 'steps', 'save_steps': 100, 'save_on_each_node': False, 'save_total_limit': None, 'enable_jit_checkpoint': False, 'push_to_hub': False, 'hub_token': '<HUB_TOKEN>', 'hub_private_repo': None, 'hub_model_id': 'w-ahmad/A-glu-waleed10-21L', 'hub_strategy': 'every_save', 'hub_always_push': False, 'hub_revision': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'restore_callback_states_from_checkpoint': False, 'full_determinism': False, 'seed': 42, 'data_seed': 42, 'use_cpu': False, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}, 'parallelism_config': None, 'dataloader_drop_last': False, 'dataloader_num_workers': 0, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'dataloader_prefetch_factor': None, 'dataloader_multiprocessing_context': None, 'dataloader_in_order': True, 'remove_unused_columns': False, 'label_names': None, 'train_sampling_strategy': 'random', 'length_column_name': 'length', 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'ddp_static_graph': None, 'ddp_backend': None, 'ddp_timeout': 1800, 'fsdp': None, 'fsdp_config': None, 'deepspeed': None, 'debug': [], 'skip_memory_metrics': True, 'do_train': False, 'do_eval': True, 'do_predict': False, 'resume_from_checkpoint': None, 'local_rank': -1}
+2026-08-14 21:31:54,415 INFO    MainThread:1229423 [wandb_config.py:__setitem__():155] [no run ID] config set model/num_parameters = 3970432 - <bound method Run._config_callback of <wandb.sdk.wandb_run.Run object at 0x1528c060ecd0>>
+2026-08-14 21:31:54,415 INFO    MainThread:1229423 [wandb_run.py:_config_callback():1346] config_cb model/num_parameters 3970432 None
+2026-08-14 21:45:48,395 INFO    MainThread:1229423 [wandb_run.py:_finish():2383] finishing run deepnevro-deepnevro/research1/q9vcmje5
+2026-08-14 21:45:48,396 INFO    MainThread:1229423 [wandb_run.py:_atexit_cleanup():2588] got exitcode: 0
+2026-08-14 21:45:48,396 INFO    MainThread:1229423 [wandb_run.py:_restore():2570] restore
+2026-08-14 21:45:48,396 INFO    MainThread:1229423 [wandb_run.py:_restore():2576] restore done
+2026-08-14 21:45:48,931 INFO    MainThread:1229423 [wandb_run.py:_footer_sync_info():3993] logging synced files

zain/Activation/wandb/run-20260814_213153-q9vcmje5/run-q9vcmje5.wandb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b17e37a7a3faacb5cb0161e4035cde89ebd95dc08857df4bc973553e68655012
+size 584687

zain/Activation/wandb/run-20260814_214549-k24y90ws/files/config.yaml

@@ -0,0 +1,439 @@
+_name_or_path:
+    value: ""
+_wandb:
+    value:
+        cli_version: 0.28.1
+        e:
+            wtedta21p9zfavbt9r6hw8vts8caym1b:
+                args:
+                    - --config
+                    - /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/configs/baseline.yaml
+                    - --variants
+                    - glu-waleed10
+                    - glu-silu-waleed10
+                    - mlp-waleed10
+                    - mlp-silu-waleed10
+                    - glu-waleed
+                    - glu-situglu_low
+                    - glu-waleedglu_low
+                codePath: sweep.py
+                codePathLocal: sweep.py
+                cpu_count: 112
+                cpu_count_logical: 224
+                cudaVersion: "12.4"
+                disk:
+                    /:
+                        total: "1560765693952"
+                        used: "709081337856"
+                email: deepnevro@gmail.com
+                executable: /mnt/data/zainulabideen/zain-exp/notebooks/my_env/bin/python
+                git:
+                    commit: 3fb47c6943d6927eafc39bd399a3d3a520bae74a
+                    remote: https://github.com/w-ahmad1a10/Activation.git
+                gpu: NVIDIA H100 80GB HBM3
+                gpu_count: 8
+                gpu_nvidia:
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-39c684a5-fde6-83d7-1663-0859795881ae
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-68012e5a-38b6-b643-0ca6-62fb66720bf3
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-132944c4-b689-2b5f-89a4-d730401677ab
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-2df386cc-6d26-d0e2-7a2d-a057b0d95864
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-bfa16575-1d94-1aa2-4537-2c93433f42ef
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-bc6c3e3c-9b90-09ca-c034-774961847c54
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-00a441e1-7c95-e7d6-4c35-43d6b291aea9
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-1c4d29a2-4647-6fce-d8fc-0c5ecfbbd6ea
+                host: deeplens-k3s-node1
+                memory:
+                    total: "2164089937920"
+                os: Linux-5.15.0-126-generic-x86_64-with-glibc2.35
+                program: /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/sweep.py
+                python: CPython 3.11.15
+                root: /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation
+                startedAt: "2026-08-14T21:45:49.461932Z"
+                writerId: wtedta21p9zfavbt9r6hw8vts8caym1b
+        m:
+            - "1": train/global_step
+              "6":
+                - 3
+              "7": []
+            - "2": '*'
+              "5": 1
+              "6":
+                - 1
+              "7": []
+        python_version: 3.11.15
+        t:
+            "1":
+                - 1
+                - 5
+                - 11
+                - 41
+                - 49
+                - 51
+                - 53
+                - 71
+            "2":
+                - 1
+                - 5
+                - 11
+                - 41
+                - 49
+                - 51
+                - 53
+                - 71
+            "3":
+                - 2
+                - 7
+                - 13
+                - 19
+                - 62
+                - 66
+            "4": 3.11.15
+            "5": 0.28.1
+            "6": 5.16.0.dev0
+            "9":
+                "1": transformers_trainer
+            "12": 0.28.1
+            "13": linux-x86_64
+accelerator_config:
+    value:
+        dispatch_batches: null
+        even_batches: true
+        gradient_accumulation_kwargs: null
+        non_blocking: false
+        split_batches: false
+        use_seedable_sampler: true
+activation:
+    value: silu-waleed10
+adam_beta1:
+    value: 0.9
+adam_beta2:
+    value: 0.999
+adam_epsilon:
+    value: 1e-08
+architectures:
+    value: null
+attention_bias:
+    value: false
+attention_dropout:
+    value: 0
+auto_find_batch_size:
+    value: false
+average_tokens_across_devices:
+    value: true
+batch_eval_metrics:
+    value: false
+bf16:
+    value: true
+bf16_full_eval:
+    value: false
+bos_token_id:
+    value: 1
+chunk_size_feed_forward:
+    value: 0
+data_seed:
+    value: 42
+dataloader_drop_last:
+    value: false
+dataloader_in_order:
+    value: true
+dataloader_multiprocessing_context:
+    value: null
+dataloader_num_workers:
+    value: 0
+dataloader_persistent_workers:
+    value: false
+dataloader_pin_memory:
+    value: true
+dataloader_prefetch_factor:
+    value: null
+ddp_backend:
+    value: null
+ddp_broadcast_buffers:
+    value: null
+ddp_bucket_cap_mb:
+    value: null
+ddp_find_unused_parameters:
+    value: null
+ddp_static_graph:
+    value: null
+ddp_timeout:
+    value: 1800
+debug:
+    value: []
+deepspeed:
+    value: null
+disable_tqdm:
+    value: false
+do_eval:
+    value: true
+do_predict:
+    value: false
+do_train:
+    value: false
+dtype:
+    value: null
+enable_jit_checkpoint:
+    value: false
+eos_token_id:
+    value: 2
+eval_accumulation_steps:
+    value: null
+eval_delay:
+    value: 0
+eval_do_concat_batches:
+    value: true
+eval_on_start:
+    value: false
+eval_steps:
+    value: 60000
+eval_strategy:
+    value: steps
+eval_use_gather_object:
+    value: false
+fp16:
+    value: false
+fp16_full_eval:
+    value: false
+fsdp:
+    value: null
+fsdp_config:
+    value: null
+full_determinism:
+    value: false
+gradient_accumulation_steps:
+    value: 4
+gradient_checkpointing:
+    value: false
+gradient_checkpointing_kwargs:
+    value: null
+greater_is_better:
+    value: null
+head_dim:
+    value: 32
+hidden_act:
+    value: silu
+hidden_size:
+    value: 128
+hub_always_push:
+    value: false
+hub_model_id:
+    value: w-ahmad/A-glu-silu-waleed10-21L
+hub_private_repo:
+    value: null
+hub_revision:
+    value: null
+hub_strategy:
+    value: every_save
+hub_token:
+    value: <HUB_TOKEN>
+id2label:
+    value:
+        "0": LABEL_0
+        "1": LABEL_1
+ignore_data_skip:
+    value: false
+include_for_metrics:
+    value: []
+include_num_input_tokens_seen:
+    value: "no"
+initializer_range:
+    value: 0.02
+intermediate_size:
+    value: 256
+is_encoder_decoder:
+    value: false
+label_names:
+    value: null
+label_smoothing_factor:
+    value: 0
+label2id:
+    value:
+        LABEL_0: 0
+        LABEL_1: 1
+learning_rate:
+    value: 0.001
+length_column_name:
+    value: length
+liger_kernel_config:
+    value: null
+load_best_model_at_end:
+    value: false
+local_rank:
+    value: -1
+log_level:
+    value: passive
+log_level_replica:
+    value: warning
+log_on_each_node:
+    value: true
+logging_first_step:
+    value: false
+logging_nan_inf_filter:
+    value: true
+logging_steps:
+    value: 20
+logging_strategy:
+    value: steps
+lr_scheduler_kwargs:
+    value: null
+lr_scheduler_type:
+    value: constant
+max_grad_norm:
+    value: 1
+max_position_embeddings:
+    value: 512
+max_steps:
+    value: 1000
+metric_for_best_model:
+    value: null
+mlp_bias:
+    value: false
+mlp_type:
+    value: glu
+model/num_parameters:
+    value: 3970432
+model_type:
+    value: tiny_llama
+neftune_noise_alpha:
+    value: null
+num_attention_heads:
+    value: 4
+num_hidden_layers:
+    value: 21
+num_key_value_heads:
+    value: 4
+num_train_epochs:
+    value: 1
+optim:
+    value: adamw_torch
+optim_args:
+    value: null
+optim_target_modules:
+    value: null
+output_attentions:
+    value: false
+output_dir:
+    value: out/glu-silu-waleed10-21L_run
+output_hidden_states:
+    value: false
+pad_token_id:
+    value: 0
+parallelism_config:
+    value: null
+per_device_eval_batch_size:
+    value: 128
+per_device_train_batch_size:
+    value: 128
+prediction_loss_only:
+    value: false
+pretraining_tp:
+    value: 1
+problem_type:
+    value: null
+project:
+    value: huggingface
+push_to_hub:
+    value: false
+remove_unused_columns:
+    value: false
+report_to:
+    value:
+        - wandb
+restore_callback_states_from_checkpoint:
+    value: false
+resume_from_checkpoint:
+    value: null
+return_dict:
+    value: true
+rms_norm_eps:
+    value: 1e-06
+rope_parameters:
+    value:
+        rope_theta: 10000
+        rope_type: default
+run_name:
+    value: LM-glu-silu-waleed10-21L-4.0M-20260814-214549
+save_on_each_node:
+    value: false
+save_only_model:
+    value: false
+save_steps:
+    value: 100
+save_strategy:
+    value: steps
+save_total_limit:
+    value: null
+seed:
+    value: 42
+skip_memory_metrics:
+    value: true
+tf32:
+    value: null
+tie_word_embeddings:
+    value: true
+tokenizer_name:
+    value: w-ahmad/tiny-stories-tokenizer
+torch_compile:
+    value: false
+torch_compile_backend:
+    value: null
+torch_compile_mode:
+    value: null
+torch_empty_cache_steps:
+    value: null
+trackio_bucket_id:
+    value: null
+trackio_space_id:
+    value: null
+trackio_static_space_id:
+    value: null
+train_sampling_strategy:
+    value: random
+transformers_version:
+    value: 5.16.0.dev0
+use_cache:
+    value: false
+use_cpu:
+    value: false
+use_liger_kernel:
+    value: false
+vocab_size:
+    value: 4096
+waleed_beta:
+    value: 10
+warmup_steps:
+    value: 0
+weight_decay:
+    value: 0.01

zain/Activation/wandb/run-20260814_214549-k24y90ws/files/output.log

@@ -0,0 +1,110 @@
+ 10%|█         | 100/1000 [01:17<11:33,  1.30it/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': '28.93', 'grad_norm': '4.031', 'learning_rate': '0.001', 'epoch': '0.01078', 'train/total_time_seconds': '7.849', 'train/time_per_step_avg': '0.3925', 'train/epoch_time_elapsed': '15.51', 'train/estimated_remaining_minutes': '6.41'}
+{'loss': '24.24', 'grad_norm': '1.328', 'learning_rate': '0.001', 'epoch': '0.02157', 'train/total_time_seconds': '15.63', 'train/time_per_step_avg': '0.3907', 'train/epoch_time_elapsed': '31.12', 'train/estimated_remaining_minutes': '6.251'}
+{'loss': '22.72', 'grad_norm': '2.844', 'learning_rate': '0.001', 'epoch': '0.03235', 'train/total_time_seconds': '23.41', 'train/time_per_step_avg': '0.3902', 'train/epoch_time_elapsed': '46.51', 'train/estimated_remaining_minutes': '6.113'}
+{'loss': '21.38', 'grad_norm': '2.031', 'learning_rate': '0.001', 'epoch': '0.04314', 'train/total_time_seconds': '31.19', 'train/time_per_step_avg': '0.3899', 'train/epoch_time_elapsed': '62.01', 'train/estimated_remaining_minutes': '5.979'}
+{'loss': '19.96', 'grad_norm': '3.125', 'learning_rate': '0.001', 'epoch': '0.05392', 'train/total_time_seconds': '38.99', 'train/time_per_step_avg': '0.3899', 'train/epoch_time_elapsed': '77.38', 'train/estimated_remaining_minutes': '5.848'}
+  - 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, 75.82it/s]
+ 20%|██        | 200/1000 [02:34<10:05,  1.32it/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': '18.43', 'grad_norm': '3.078', 'learning_rate': '0.001', 'epoch': '0.06471', 'train/total_time_seconds': '46.81', 'train/time_per_step_avg': '0.3896', 'train/epoch_time_elapsed': '92.97', 'train/estimated_remaining_minutes': '5.721'}
+{'loss': '17.22', 'grad_norm': '3.656', 'learning_rate': '0.001', 'epoch': '0.07549', 'train/total_time_seconds': '54.61', 'train/time_per_step_avg': '0.3898', 'train/epoch_time_elapsed': '108.3', 'train/estimated_remaining_minutes': '5.591'}
+{'loss': '16.27', 'grad_norm': '3.156', 'learning_rate': '0.001', 'epoch': '0.08628', 'train/total_time_seconds': '62.38', 'train/time_per_step_avg': '0.3896', 'train/epoch_time_elapsed': '124.1', 'train/estimated_remaining_minutes': '5.458'}
+{'loss': '15.49', 'grad_norm': '2.359', 'learning_rate': '0.001', 'epoch': '0.09706', 'train/total_time_seconds': '70.13', 'train/time_per_step_avg': '0.3894', 'train/epoch_time_elapsed': '139.3', 'train/estimated_remaining_minutes': '5.325'}
+{'loss': '14.8', 'grad_norm': '2.797', 'learning_rate': '0.001', 'epoch': '0.1078', 'train/total_time_seconds': '77.89', 'train/time_per_step_avg': '0.3891', 'train/epoch_time_elapsed': '154.6', 'train/estimated_remaining_minutes': '5.193'}
+  - 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, 79.86it/s]
+ 30%|███       | 300/1000 [03:51<08:52,  1.31it/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': '14.22', 'grad_norm': '2.25', 'learning_rate': '0.001', 'epoch': '0.1186', 'train/total_time_seconds': '85.67', 'train/time_per_step_avg': '0.3887', 'train/epoch_time_elapsed': '170.1', 'train/estimated_remaining_minutes': '5.062'}
+{'loss': '13.73', 'grad_norm': '2.641', 'learning_rate': '0.001', 'epoch': '0.1294', 'train/total_time_seconds': '93.44', 'train/time_per_step_avg': '0.3883', 'train/epoch_time_elapsed': '185.5', 'train/estimated_remaining_minutes': '4.931'}
+{'loss': '13.25', 'grad_norm': '2.172', 'learning_rate': '0.001', 'epoch': '0.1402', 'train/total_time_seconds': '101.2', 'train/time_per_step_avg': '0.3881', 'train/epoch_time_elapsed': '200.7', 'train/estimated_remaining_minutes': '4.8'}
+{'loss': '12.85', 'grad_norm': '1.781', 'learning_rate': '0.001', 'epoch': '0.151', 'train/total_time_seconds': '109', 'train/time_per_step_avg': '0.3886', 'train/epoch_time_elapsed': '216.3', 'train/estimated_remaining_minutes': '4.671'}
+{'loss': '12.48', 'grad_norm': '2.5', 'learning_rate': '0.001', 'epoch': '0.1618', 'train/total_time_seconds': '116.8', 'train/time_per_step_avg': '0.3888', 'train/epoch_time_elapsed': '231.6', 'train/estimated_remaining_minutes': '4.541'}
+  - 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, 52.93it/s]
+ 40%|████      | 400/1000 [05:08<07:39,  1.31it/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': '12.14', 'grad_norm': '3.062', 'learning_rate': '0.001', 'epoch': '0.1726', 'train/total_time_seconds': '124.5', 'train/time_per_step_avg': '0.3886', 'train/epoch_time_elapsed': '246.9', 'train/estimated_remaining_minutes': '4.41'}
+{'loss': '11.86', 'grad_norm': '2.359', 'learning_rate': '0.001', 'epoch': '0.1833', 'train/total_time_seconds': '132.3', 'train/time_per_step_avg': '0.3884', 'train/epoch_time_elapsed': '262.1', 'train/estimated_remaining_minutes': '4.279'}
+{'loss': '11.56', 'grad_norm': '2.516', 'learning_rate': '0.001', 'epoch': '0.1941', 'train/total_time_seconds': '140.1', 'train/time_per_step_avg': '0.3888', 'train/epoch_time_elapsed': '277.5', 'train/estimated_remaining_minutes': '4.15'}
+{'loss': '11.3', 'grad_norm': '2.234', 'learning_rate': '0.001', 'epoch': '0.2049', 'train/total_time_seconds': '147.9', 'train/time_per_step_avg': '0.3887', 'train/epoch_time_elapsed': '292.9', 'train/estimated_remaining_minutes': '4.021'}
+{'loss': '11.09', 'grad_norm': '1.984', 'learning_rate': '0.001', 'epoch': '0.2157', 'train/total_time_seconds': '155.7', 'train/time_per_step_avg': '0.3888', 'train/epoch_time_elapsed': '308.3', 'train/estimated_remaining_minutes': '3.891'}
+  - 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, 77.32it/s]
+ 50%|█████     | 500/1000 [06:25<06:19,  1.32it/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': '10.86', 'grad_norm': '3.016', 'learning_rate': '0.001', 'epoch': '0.2265', 'train/total_time_seconds': '163.5', 'train/time_per_step_avg': '0.3893', 'train/epoch_time_elapsed': '323.7', 'train/estimated_remaining_minutes': '3.762'}
+{'loss': '10.72', 'grad_norm': '2.234', 'learning_rate': '0.001', 'epoch': '0.2373', 'train/total_time_seconds': '171.2', 'train/time_per_step_avg': '0.3896', 'train/epoch_time_elapsed': '339.4', 'train/estimated_remaining_minutes': '3.632'}
+{'loss': '10.52', 'grad_norm': '2.328', 'learning_rate': '0.001', 'epoch': '0.248', 'train/total_time_seconds': '179', 'train/time_per_step_avg': '0.3892', 'train/epoch_time_elapsed': '354.7', 'train/estimated_remaining_minutes': '3.502'}
+{'loss': '10.35', 'grad_norm': '2.938', 'learning_rate': '0.001', 'epoch': '0.2588', 'train/total_time_seconds': '186.7', 'train/time_per_step_avg': '0.3889', 'train/epoch_time_elapsed': '370', 'train/estimated_remaining_minutes': '3.372'}
+{'loss': '10.2', 'grad_norm': '2.438', 'learning_rate': '0.001', 'epoch': '0.2696', 'train/total_time_seconds': '194.5', 'train/time_per_step_avg': '0.3885', 'train/epoch_time_elapsed': '385.3', 'train/estimated_remaining_minutes': '3.242'}
+  - 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, 81.80it/s]
+ 60%|██████    | 600/1000 [07:42<05:09,  1.29it/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': '10.09', 'grad_norm': '2.25', 'learning_rate': '0.001', 'epoch': '0.2804', 'train/total_time_seconds': '202.3', 'train/time_per_step_avg': '0.3884', 'train/epoch_time_elapsed': '400.9', 'train/estimated_remaining_minutes': '3.112'}
+{'loss': '9.965', 'grad_norm': '2', 'learning_rate': '0.001', 'epoch': '0.2912', 'train/total_time_seconds': '210.1', 'train/time_per_step_avg': '0.3886', 'train/epoch_time_elapsed': '416.3', 'train/estimated_remaining_minutes': '2.983'}
+{'loss': '9.853', 'grad_norm': '1.992', 'learning_rate': '0.001', 'epoch': '0.302', 'train/total_time_seconds': '217.9', 'train/time_per_step_avg': '0.389', 'train/epoch_time_elapsed': '431.8', 'train/estimated_remaining_minutes': '2.853'}
+{'loss': '9.736', 'grad_norm': '1.852', 'learning_rate': '0.001', 'epoch': '0.3128', 'train/total_time_seconds': '225.7', 'train/time_per_step_avg': '0.3895', 'train/epoch_time_elapsed': '447.2', 'train/estimated_remaining_minutes': '2.724'}
+{'loss': '9.628', 'grad_norm': '2.656', 'learning_rate': '0.001', 'epoch': '0.3235', 'train/total_time_seconds': '233.5', 'train/time_per_step_avg': '0.3899', 'train/epoch_time_elapsed': '462.9', 'train/estimated_remaining_minutes': '2.594'}
+  - 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, 79.34it/s]
+ 70%|███████   | 700/1000 [08:59<03:50,  1.30it/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': '9.536', 'grad_norm': '2.25', 'learning_rate': '0.001', 'epoch': '0.3343', 'train/total_time_seconds': '241.3', 'train/time_per_step_avg': '0.3897', 'train/epoch_time_elapsed': '478.3', 'train/estimated_remaining_minutes': '2.465'}
+{'loss': '9.441', 'grad_norm': '2.031', 'learning_rate': '0.001', 'epoch': '0.3451', 'train/total_time_seconds': '249', 'train/time_per_step_avg': '0.3893', 'train/epoch_time_elapsed': '493.7', 'train/estimated_remaining_minutes': '2.335'}
+{'loss': '9.348', 'grad_norm': '2.219', 'learning_rate': '0.001', 'epoch': '0.3559', 'train/total_time_seconds': '256.8', 'train/time_per_step_avg': '0.389', 'train/epoch_time_elapsed': '509.1', 'train/estimated_remaining_minutes': '2.205'}
+{'loss': '9.265', 'grad_norm': '2.234', 'learning_rate': '0.001', 'epoch': '0.3667', 'train/total_time_seconds': '264.6', 'train/time_per_step_avg': '0.3889', 'train/epoch_time_elapsed': '524.6', 'train/estimated_remaining_minutes': '2.075'}
+{'loss': '9.19', 'grad_norm': '1.641', 'learning_rate': '0.001', 'epoch': '0.3775', 'train/total_time_seconds': '272.4', 'train/time_per_step_avg': '0.3888', 'train/epoch_time_elapsed': '540', 'train/estimated_remaining_minutes': '1.945'}
+  - 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, 46.72it/s]
+ 80%|████████  | 800/1000 [10:17<02:35,  1.29it/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': '9.1', 'grad_norm': '2.156', 'learning_rate': '0.001', 'epoch': '0.3882', 'train/total_time_seconds': '280.1', 'train/time_per_step_avg': '0.3887', 'train/epoch_time_elapsed': '555.5', 'train/estimated_remaining_minutes': '1.816'}
+{'loss': '9.055', 'grad_norm': '2.531', 'learning_rate': '0.001', 'epoch': '0.399', 'train/total_time_seconds': '287.9', 'train/time_per_step_avg': '0.3888', 'train/epoch_time_elapsed': '570.9', 'train/estimated_remaining_minutes': '1.686'}
+{'loss': '8.981', 'grad_norm': '2.016', 'learning_rate': '0.001', 'epoch': '0.4098', 'train/total_time_seconds': '295.7', 'train/time_per_step_avg': '0.3889', 'train/epoch_time_elapsed': '586.4', 'train/estimated_remaining_minutes': '1.556'}
+{'loss': '8.936', 'grad_norm': '2.359', 'learning_rate': '0.001', 'epoch': '0.4206', 'train/total_time_seconds': '303.5', 'train/time_per_step_avg': '0.3889', 'train/epoch_time_elapsed': '601.8', 'train/estimated_remaining_minutes': '1.427'}
+{'loss': '8.839', 'grad_norm': '2.156', 'learning_rate': '0.001', 'epoch': '0.4314', 'train/total_time_seconds': '311.3', 'train/time_per_step_avg': '0.3893', 'train/epoch_time_elapsed': '617.3', 'train/estimated_remaining_minutes': '1.297'}
+  - 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, 77.94it/s]
+ 90%|█████████ | 900/1000 [11:34<01:16,  1.30it/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': '8.804', 'grad_norm': '2.344', 'learning_rate': '0.001', 'epoch': '0.4422', 'train/total_time_seconds': '319.1', 'train/time_per_step_avg': '0.3893', 'train/epoch_time_elapsed': '632.8', 'train/estimated_remaining_minutes': '1.167'}
+{'loss': '8.767', 'grad_norm': '2.281', 'learning_rate': '0.001', 'epoch': '0.453', 'train/total_time_seconds': '326.9', 'train/time_per_step_avg': '0.3898', 'train/epoch_time_elapsed': '648.4', 'train/estimated_remaining_minutes': '1.038'}
+{'loss': '8.691', 'grad_norm': '2.438', 'learning_rate': '0.001', 'epoch': '0.4637', 'train/total_time_seconds': '334.7', 'train/time_per_step_avg': '0.3898', 'train/epoch_time_elapsed': '663.8', 'train/estimated_remaining_minutes': '0.908'}
+{'loss': '8.652', 'grad_norm': '1.828', 'learning_rate': '0.001', 'epoch': '0.4745', 'train/total_time_seconds': '342.4', 'train/time_per_step_avg': '0.3895', 'train/epoch_time_elapsed': '679.2', 'train/estimated_remaining_minutes': '0.7782'}
+{'loss': '8.596', 'grad_norm': '2.5', 'learning_rate': '0.001', 'epoch': '0.4853', 'train/total_time_seconds': '350.2', 'train/time_per_step_avg': '0.3895', 'train/epoch_time_elapsed': '694.6', 'train/estimated_remaining_minutes': '0.6486'}
+  - 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, 76.93it/s]
+100%|██████████| 1000/1000 [13:01<00:00,  1.30i[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': '8.548', 'grad_norm': '2.25', 'learning_rate': '0.001', 'epoch': '0.4961', 'train/total_time_seconds': '358', 'train/time_per_step_avg': '0.3897', 'train/epoch_time_elapsed': '710.2', 'train/estimated_remaining_minutes': '0.5189'}
+{'loss': '8.527', 'grad_norm': '2.188', 'learning_rate': '0.001', 'epoch': '0.5069', 'train/total_time_seconds': '365.9', 'train/time_per_step_avg': '0.3897', 'train/epoch_time_elapsed': '725.5', 'train/estimated_remaining_minutes': '0.3892'}
+{'loss': '8.503', 'grad_norm': '1.906', 'learning_rate': '0.001', 'epoch': '0.5177', 'train/total_time_seconds': '373.7', 'train/time_per_step_avg': '0.39', 'train/epoch_time_elapsed': '741', 'train/estimated_remaining_minutes': '0.2595'}
+{'loss': '8.467', 'grad_norm': '2.141', 'learning_rate': '0.001', 'epoch': '0.5284', 'train/total_time_seconds': '381.5', 'train/time_per_step_avg': '0.3903', 'train/epoch_time_elapsed': '756.3', 'train/estimated_remaining_minutes': '0.1297'}
+{'loss': '8.423', 'grad_norm': '2', 'learning_rate': '0.001', 'epoch': '0.5392', 'train/total_time_seconds': '389.3', 'train/time_per_step_avg': '0.3902', 'train/epoch_time_elapsed': '771.8', 'train/estimated_remaining_minutes': '0'}
+  - 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': '2.102', 'eval_runtime': '9.474', 'eval_samples_per_second': '1006', 'eval_steps_per_second': '7.916', 'epoch': '0.5392', 'train/total_time_seconds': '389.3', 'train/time_per_step_avg': '0.3902', 'train/epoch_time_elapsed': '781.3', 'train/estimated_remaining_minutes': '0'}
+  - 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, 79.01it/s]
+100%|██████████| 1000/1000 [13:01<00:00,  1.28it/s], ?it/s]
+{'train_runtime': '782.2', 'train_samples_per_second': '654.5', 'train_steps_per_second': '1.278', 'train_loss': '12.07', 'epoch': '0.5392', 'train/total_time_seconds': '389.3', 'train/time_per_step_avg': '0.3902', 'train/epoch_time_elapsed': '781.3', 'train/estimated_remaining_minutes': '0'}
+100%|██████████| 75/75 [00:09<00:00,  8.08it/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.
+  - 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, 81.19it/s]
+>>> FINISHED glu-silu-waleed10-21L successfully

zain/Activation/wandb/run-20260814_214549-k24y90ws/files/requirements.txt

@@ -0,0 +1,149 @@
+asttokens==3.0.1
+comm==0.2.3
+debugpy==1.8.21
+decorator==5.3.1
+executing==2.2.1
+nest-asyncio==1.6.0
+parso==0.8.7
+platformdirs==4.11.0
+psutil==7.2.2
+ptyprocess==0.7.0
+pure_eval==0.2.3
+Pygments==2.20.0
+pyzmq==27.1.0
+setuptools==83.0.0
+six==1.17.0
+tornado==6.5.7
+traitlets==5.15.0
+fsspec==2026.4.0
+wcwidth==0.8.2
+ipython_pygments_lexers==1.1.1
+jedi==0.20.0
+jupyter_core==5.9.1
+matplotlib-inline==0.2.2
+pexpect==4.9.0
+prompt_toolkit==3.0.53
+python-dateutil==2.9.0.post0
+stack_data==0.6.3
+wheel==0.47.0
+jupyter_client==8.9.1
+pip==26.1.2
+ipython==9.15.0
+ipykernel==7.2.0
+threadpoolctl==3.6.0
+pyparsing==3.3.2
+typing_extensions==4.15.0
+Jinja2==3.1.6
+narwhals==2.24.0
+kiwisolver==1.5.0
+joblib==1.5.3
+fonttools==4.63.0
+cycler==0.12.1
+scipy==1.17.1
+pandas==3.0.5
+contourpy==1.3.3
+scikit-learn==1.9.0
+matplotlib==3.11.1
+urllib3==2.7.0
+tqdm==4.70.0
+idna==3.18
+charset-normalizer==3.4.9
+certifi==2026.7.22
+requests==2.34.2
+seaborn==0.13.2
+uv==0.12.0
+shellingham==1.5.4
+mpmath==1.3.0
+attrs==26.1.0
+hf-xet==1.5.2
+nvidia-nccl-cu12==2.21.5
+MarkupSafe==3.0.3
+regex==2026.7.19
+importlib_metadata==9.0.0
+httpcore==1.0.9
+annotated-doc==0.0.5
+multidict==6.7.1
+aiohttp==3.14.3
+aiosignal==1.4.0
+xxhash==3.8.1
+aiohappyeyeballs==2.7.1
+mdurl==0.1.2
+cuda-toolkit==13.0.3.0
+networkx==3.6.1
+PyYAML==6.0.3
+nvidia-cufile==1.15.1.6
+typer==0.27.0
+torchaudio==2.6.0+cu124
+rich==15.0.0
+nvidia-cufft-cu12==11.2.1.3
+h11==0.16.0
+dill==0.4.1
+cuda-pathfinder==1.6.0
+filelock==3.29.0
+nvidia-nvtx-cu12==12.4.127
+httpx==0.28.1
+anyio==4.14.2
+numpy==2.4.4
+yarl==1.24.5
+click==8.4.2
+triton==3.2.0
+frozenlist==1.8.0
+zipp==4.1.0
+propcache==0.5.2
+tokenizers==0.22.2
+markdown-it-py==4.2.0
+nvidia-cuda-runtime==13.0.96
+cuda-bindings==13.3.1
+nvidia-cuda-cupti==13.0.85
+torch==2.6.0+cu124
+multiprocess==0.70.19
+pillow==12.2.0
+transformers==5.16.0.dev0
+wandb==0.28.1
+nvidia-curand==10.4.0.35
+sympy==1.13.1
+nvidia-cusparse==12.6.3.3
+nvidia-cuda-nvrtc==13.0.88
+typing-inspection==0.4.2
+nvidia-cusolver==12.0.4.66
+nvidia-cufft==12.0.0.61
+nvidia-cudnn-cu13==9.20.0.48
+nvidia-cublas==13.1.1.3
+pyarrow==25.0.0
+evaluate==0.4.6
+diffusers==0.39.0
+pydantic==2.13.4
+annotated-types==0.8.0
+protobuf==7.35.1
+sentry-sdk==2.66.1
+einops==0.8.2
+packaging==26.2
+nvidia-nvjitlink-cu12==12.4.127
+nvidia-curand-cu12==10.3.5.147
+nvidia-cusparselt-cu12==0.6.2
+nvidia-cusparse-cu12==12.3.1.170
+nvidia-cuda-runtime-cu12==12.4.127
+torchvision==0.21.0+cu124
+nvidia-cuda-nvrtc-cu12==12.4.127
+nvidia-cuda-cupti-cu12==12.4.127
+nvidia-cusolver-cu12==11.6.1.9
+nvidia-cublas-cu12==12.4.5.8
+nvidia-cudnn-cu12==9.1.0.70
+huggingface_hub==1.26.0
+datasets==5.0.1
+safetensors==0.8.0
+accelerate==1.14.0
+pydantic_core==2.46.4
+ninja==1.13.0
+autocommand==2.2.2
+backports.tarfile==1.2.0
+importlib_metadata==8.7.1
+jaraco.text==4.0.0
+jaraco.context==6.1.0
+jaraco.functools==4.4.0
+more-itertools==10.8.0
+packaging==26.0
+platformdirs==4.4.0
+tomli==2.4.0
+wheel==0.46.3
+zipp==3.23.0

zain/Activation/wandb/run-20260814_214549-k24y90ws/files/wandb-metadata.json

@@ -0,0 +1,101 @@
+{
+  "os":  "Linux-5.15.0-126-generic-x86_64-with-glibc2.35",
+  "python":  "CPython 3.11.15",
+  "startedAt":  "2026-08-14T21:45:49.461932Z",
+  "args":  [
+    "--config",
+    "/mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/configs/baseline.yaml",
+    "--variants",
+    "glu-waleed10",
+    "glu-silu-waleed10",
+    "mlp-waleed10",
+    "mlp-silu-waleed10",
+    "glu-waleed",
+    "glu-situglu_low",
+    "glu-waleedglu_low"
+  ],
+  "program":  "/mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/sweep.py",
+  "codePath":  "sweep.py",
+  "codePathLocal":  "sweep.py",
+  "git":  {
+    "remote":  "https://github.com/w-ahmad1a10/Activation.git",
+    "commit":  "3fb47c6943d6927eafc39bd399a3d3a520bae74a"
+  },
+  "email":  "deepnevro@gmail.com",
+  "root":  "/mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation",
+  "host":  "deeplens-k3s-node1",
+  "executable":  "/mnt/data/zainulabideen/zain-exp/notebooks/my_env/bin/python",
+  "cpu_count":  112,
+  "cpu_count_logical":  224,
+  "gpu":  "NVIDIA H100 80GB HBM3",
+  "gpu_count":  8,
+  "disk":  {
+    "/":  {
+      "total":  "1560765693952",
+      "used":  "709081337856"
+    }
+  },
+  "memory":  {
+    "total":  "2164089937920"
+  },
+  "gpu_nvidia":  [
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-39c684a5-fde6-83d7-1663-0859795881ae"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-68012e5a-38b6-b643-0ca6-62fb66720bf3"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-132944c4-b689-2b5f-89a4-d730401677ab"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-2df386cc-6d26-d0e2-7a2d-a057b0d95864"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-bfa16575-1d94-1aa2-4537-2c93433f42ef"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-bc6c3e3c-9b90-09ca-c034-774961847c54"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-00a441e1-7c95-e7d6-4c35-43d6b291aea9"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-1c4d29a2-4647-6fce-d8fc-0c5ecfbbd6ea"
+    }
+  ],
+  "cudaVersion":  "12.4",
+  "writerId":  "wtedta21p9zfavbt9r6hw8vts8caym1b"
+}

zain/Activation/wandb/run-20260814_214549-k24y90ws/files/wandb-summary.json

@@ -0,0 +1 @@
+{"train_samples_per_second":654.524,"eval/steps_per_second":7.91,"train_steps_per_second":1.278,"_runtime":790,"total_flos":5.420315836416e+15,"_step":52,"train_loss":12.069926330566407,"train/grad_norm":2,"eval/runtime":9.4813,"_timestamp":1.7867447411310596e+09,"train/global_step":1000,"train_runtime":782.2476,"eval/samples_per_second":1004.817,"train/epoch":0.5392289026691831,"train/learning_rate":0.001,"_wandb":{"runtime":790},"train/loss":8.422817993164063,"eval/loss":2.101808547973633}

zain/Activation/wandb/run-20260814_214549-k24y90ws/logs/debug-core.log

@@ -0,0 +1,62 @@
+{"time":"2026-08-14T21:31:53.539152008Z","level":"INFO","msg":"main: starting server","port-filename":"/tmp/tmp650y3sc2/port-1229423.txt","pid":1229423,"detached":false,"idle-timeout":600000000000,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false,"enable-dcgm-profiling":false}
+{"time":"2026-08-14T21:31:53.540676876Z","level":"INFO","msg":"server: will exit if parent process dies","ppid":1229423}
+{"time":"2026-08-14T21:31:53.540659035Z","level":"INFO","msg":"server: accepting connections","addr":{"Name":"/tmp/wandb-1229423-1230766-2604858008/socket","Net":"unix"}}
+{"time":"2026-08-14T21:31:53.717724383Z","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"1(@)"}
+{"time":"2026-08-14T21:31:53.802092564Z","level":"INFO","msg":"handleInformInit: received","streamId":"q9vcmje5","id":"1(@)"}
+{"time":"2026-08-14T21:31:54.062397411Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"q9vcmje5","id":"1(@)"}
+{"time":"2026-08-14T21:31:59.411924652Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"32s1adqk73rr"}
+{"time":"2026-08-14T21:45:48.396635188Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"32s1adqk73rr"}
+{"time":"2026-08-14T21:45:48.929834053Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"32s1adqk73rr"}
+{"time":"2026-08-14T21:45:48.931620539Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"q9vcmje5","id":"1(@)"}
+{"time":"2026-08-14T21:45:48.932491197Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"q9vcmje5","id":"1(@)"}
+{"time":"2026-08-14T21:45:49.464388506Z","level":"INFO","msg":"handleInformInit: received","streamId":"k24y90ws","id":"1(@)"}
+{"time":"2026-08-14T21:45:49.733853364Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"k24y90ws","id":"1(@)"}
+{"time":"2026-08-14T21:45:55.329792921Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"dr322rnt7xgx"}
+{"time":"2026-08-14T21:59:01.158537572Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"dr322rnt7xgx"}
+{"time":"2026-08-14T21:59:01.914569559Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"dr322rnt7xgx"}
+{"time":"2026-08-14T21:59:01.916156477Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"k24y90ws","id":"1(@)"}
+{"time":"2026-08-14T21:59:01.917534874Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"k24y90ws","id":"1(@)"}
+{"time":"2026-08-14T21:59:02.524868682Z","level":"INFO","msg":"handleInformInit: received","streamId":"npap5k3g","id":"1(@)"}
+{"time":"2026-08-14T21:59:02.786199875Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"npap5k3g","id":"1(@)"}
+{"time":"2026-08-14T21:59:08.192858634Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"2g99yy5pussw"}
+{"time":"2026-08-14T22:11:53.643379161Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"2g99yy5pussw"}
+{"time":"2026-08-14T22:11:54.173579992Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"2g99yy5pussw"}
+{"time":"2026-08-14T22:11:54.17490129Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"npap5k3g","id":"1(@)"}
+{"time":"2026-08-14T22:11:54.175479186Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"npap5k3g","id":"1(@)"}
+{"time":"2026-08-14T22:11:54.694802218Z","level":"INFO","msg":"handleInformInit: received","streamId":"a3607q94","id":"1(@)"}
+{"time":"2026-08-14T22:11:54.956642682Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"a3607q94","id":"1(@)"}
+{"time":"2026-08-14T22:12:00.441973218Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"raaqqqmajsfs"}
+{"time":"2026-08-14T22:24:55.662446346Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"raaqqqmajsfs"}
+{"time":"2026-08-14T22:24:56.296327526Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"raaqqqmajsfs"}
+{"time":"2026-08-14T22:24:56.297810539Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"a3607q94","id":"1(@)"}
+{"time":"2026-08-14T22:24:56.29856977Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"a3607q94","id":"1(@)"}
+{"time":"2026-08-14T22:24:56.82560196Z","level":"INFO","msg":"handleInformInit: received","streamId":"wcvy5vyw","id":"1(@)"}
+{"time":"2026-08-14T22:24:57.085772231Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"wcvy5vyw","id":"1(@)"}
+{"time":"2026-08-14T22:25:02.571913681Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"2nb8wusqu4fo"}
+{"time":"2026-08-14T22:38:28.612623637Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"2nb8wusqu4fo"}
+{"time":"2026-08-14T22:38:29.208727987Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"2nb8wusqu4fo"}
+{"time":"2026-08-14T22:38:29.210106262Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"wcvy5vyw","id":"1(@)"}
+{"time":"2026-08-14T22:38:29.210705504Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"wcvy5vyw","id":"1(@)"}
+{"time":"2026-08-14T22:38:29.839594174Z","level":"INFO","msg":"handleInformInit: received","streamId":"uuv4h936","id":"1(@)"}
+{"time":"2026-08-14T22:38:30.099444184Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"uuv4h936","id":"1(@)"}
+{"time":"2026-08-14T22:38:35.620666929Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"r0aouorh27ls"}
+{"time":"2026-08-14T22:52:17.704425638Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"r0aouorh27ls"}
+{"time":"2026-08-14T22:52:18.35448297Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"r0aouorh27ls"}
+{"time":"2026-08-14T22:52:18.355720686Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"uuv4h936","id":"1(@)"}
+{"time":"2026-08-14T22:52:18.356578583Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"uuv4h936","id":"1(@)"}
+{"time":"2026-08-14T22:52:18.899799881Z","level":"INFO","msg":"handleInformInit: received","streamId":"x87hwrtd","id":"1(@)"}
+{"time":"2026-08-14T22:52:19.159494071Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"x87hwrtd","id":"1(@)"}
+{"time":"2026-08-14T22:52:24.671083701Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"hn7lqjhi3ekb"}
+{"time":"2026-08-14T23:05:51.214192173Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"hn7lqjhi3ekb"}
+{"time":"2026-08-14T23:05:51.783107435Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"hn7lqjhi3ekb"}
+{"time":"2026-08-14T23:05:51.784410941Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"x87hwrtd","id":"1(@)"}
+{"time":"2026-08-14T23:05:51.78516667Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"x87hwrtd","id":"1(@)"}
+{"time":"2026-08-14T23:05:52.088192277Z","level":"INFO","msg":"handleInformTeardown: server teardown initiated","id":"1(@)"}
+{"time":"2026-08-14T23:05:52.088229648Z","level":"INFO","msg":"handleInformTeardown: server shutdown complete","id":"1(@)"}
+{"time":"2026-08-14T23:05:52.088239651Z","level":"INFO","msg":"connection: closing","id":"1(@)"}
+{"time":"2026-08-14T23:05:52.088243873Z","level":"INFO","msg":"server: is shutting down"}
+{"time":"2026-08-14T23:05:52.0882852Z","level":"INFO","msg":"connection: closed successfully","id":"1(@)"}
+{"time":"2026-08-14T23:05:52.088299095Z","level":"INFO","msg":"processOutgoingData: finished","id":"1(@)"}
+{"time":"2026-08-14T23:05:52.088345917Z","level":"INFO","msg":"connection: ManageConnectionData: connection closed","id":"1(@)"}
+{"time":"2026-08-14T23:05:52.088452051Z","level":"INFO","msg":"server: listener closed","addr":{"Name":"/tmp/wandb-1229423-1230766-2604858008/socket","Net":"unix"}}
+{"time":"2026-08-14T23:05:52.088520571Z","level":"INFO","msg":"server: all connections closed"}

zain/Activation/wandb/run-20260814_214549-k24y90ws/logs/debug-internal.log

@@ -0,0 +1,123 @@
+{"time":"2026-08-14T21:45:49.464511458Z","level":"INFO","msg":"wandb-core"}
+{"time":"2026-08-14T21:45:49.464655404Z","level":"INFO","msg":"stream: starting","core version":"0.28.1"}
+{"time":"2026-08-14T21:45:49.733697059Z","level":"INFO","msg":"stream: created new stream","id":"k24y90ws"}
+{"time":"2026-08-14T21:45:49.733775349Z","level":"INFO","msg":"handler: started"}
+{"time":"2026-08-14T21:45:49.733847443Z","level":"INFO","msg":"stream: started"}
+{"time":"2026-08-14T21:45:49.733855982Z","level":"INFO","msg":"writer: started","stream_id":"k24y90ws"}
+{"time":"2026-08-14T21:45:49.733908797Z","level":"INFO","msg":"sender: started"}
+{"time":"2026-08-14T21:45:50.320642056Z","level":"INFO","msg":"filestream: sending request","total_files":1,"console_offset":0,"console_lines":1}
+{"time":"2026-08-14T21:45:50.446758607Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:46:05.320807587Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":0,"events_lines":1,"console_offset":0,"console_lines":1,"uploaded_len":2}
+{"time":"2026-08-14T21:46:05.495295275Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:46:20.321362864Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":0,"history_lines":1,"events_offset":1,"events_lines":2,"console_offset":0,"console_lines":2}
+{"time":"2026-08-14T21:46:20.588211042Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:46:35.321170502Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":1,"history_lines":1,"events_offset":3,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-14T21:46:35.523653777Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:46:50.321411954Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":2,"history_lines":1,"events_offset":5,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-14T21:46:50.578357633Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:47:05.320973253Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":3,"history_lines":1,"events_offset":7,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-14T21:47:05.494256218Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:47:20.321280351Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":4,"history_lines":1,"events_offset":9,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-14T21:47:20.471515967Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:47:35.321022569Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":5,"history_lines":1,"events_offset":11,"events_lines":2,"console_offset":2,"console_lines":10}
+{"time":"2026-08-14T21:47:35.496353897Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:47:50.320853074Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":6,"history_lines":1,"events_offset":13,"events_lines":2,"console_offset":10,"console_lines":1}
+{"time":"2026-08-14T21:47:50.483129663Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:48:05.32118415Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":7,"history_lines":1,"events_offset":15,"events_lines":2,"console_offset":10,"console_lines":1}
+{"time":"2026-08-14T21:48:05.503672796Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:48:20.321043028Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":8,"history_lines":1,"events_offset":17,"events_lines":2,"console_offset":10,"console_lines":1}
+{"time":"2026-08-14T21:48:20.530090466Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:48:35.321781731Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":9,"history_lines":1,"events_offset":19,"events_lines":2,"console_offset":10,"console_lines":1}
+{"time":"2026-08-14T21:48:35.508321767Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:48:50.321120237Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":10,"history_lines":1,"events_offset":21,"events_lines":2,"console_offset":12,"console_lines":10}
+{"time":"2026-08-14T21:48:50.442521449Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:49:05.320862907Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":11,"history_lines":1,"events_offset":23,"events_lines":2,"console_offset":20,"console_lines":1}
+{"time":"2026-08-14T21:49:05.535886019Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:49:20.321534303Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":12,"history_lines":1,"events_offset":25,"events_lines":2,"console_offset":20,"console_lines":1}
+{"time":"2026-08-14T21:49:20.473500807Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:49:35.320974561Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":13,"history_lines":1,"events_offset":27,"events_lines":2,"console_offset":20,"console_lines":1}
+{"time":"2026-08-14T21:49:35.452315273Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:49:50.321583693Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":14,"history_lines":1,"events_offset":29,"events_lines":2,"console_offset":20,"console_lines":1}
+{"time":"2026-08-14T21:49:50.475924855Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:50:05.321095412Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":15,"history_lines":1,"events_offset":31,"events_lines":2,"console_offset":22,"console_lines":10}
+{"time":"2026-08-14T21:50:05.510373585Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:50:20.321072924Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":16,"history_lines":1,"events_offset":33,"events_lines":2,"console_offset":30,"console_lines":1}
+{"time":"2026-08-14T21:50:20.441403714Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:50:35.320763733Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":17,"history_lines":1,"events_offset":35,"events_lines":2,"console_offset":30,"console_lines":1}
+{"time":"2026-08-14T21:50:35.466119892Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:50:50.321438704Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":18,"history_lines":1,"events_offset":37,"events_lines":2,"console_offset":30,"console_lines":1}
+{"time":"2026-08-14T21:50:50.456280209Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:51:05.320761985Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":19,"history_lines":1,"events_offset":39,"events_lines":2,"console_offset":30,"console_lines":1}
+{"time":"2026-08-14T21:51:05.477583946Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:51:20.321080069Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":20,"history_lines":1,"events_offset":41,"events_lines":2,"console_offset":32,"console_lines":10}
+{"time":"2026-08-14T21:51:20.453021894Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:51:35.320796936Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":21,"history_lines":1,"events_offset":43,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-14T21:51:35.466525803Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:51:50.320955591Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":22,"history_lines":1,"events_offset":45,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-14T21:51:50.469319734Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:52:05.321309506Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":23,"history_lines":1,"events_offset":47,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-14T21:52:05.501402407Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:52:20.321650872Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":24,"history_lines":1,"events_offset":49,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-14T21:52:20.4534208Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:52:35.32088535Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":25,"history_lines":1,"events_offset":51,"events_lines":2,"console_offset":42,"console_lines":10}
+{"time":"2026-08-14T21:52:35.461275831Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:52:50.321176023Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":26,"history_lines":1,"events_offset":53,"events_lines":2,"console_offset":50,"console_lines":1}
+{"time":"2026-08-14T21:52:50.466826609Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:53:05.321699021Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":27,"history_lines":1,"events_offset":55,"events_lines":2,"console_offset":50,"console_lines":1}
+{"time":"2026-08-14T21:53:05.486511079Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:53:20.321112129Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":28,"history_lines":1,"events_offset":57,"events_lines":2,"console_offset":50,"console_lines":1}
+{"time":"2026-08-14T21:53:20.457369802Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:53:35.320797688Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":29,"history_lines":1,"events_offset":59,"events_lines":2,"console_offset":50,"console_lines":1}
+{"time":"2026-08-14T21:53:35.452345981Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:53:50.321742882Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":30,"history_lines":1,"events_offset":61,"events_lines":2,"console_offset":52,"console_lines":10}
+{"time":"2026-08-14T21:53:50.434938647Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:54:05.321429864Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":31,"history_lines":1,"events_offset":63,"events_lines":2,"console_offset":60,"console_lines":1}
+{"time":"2026-08-14T21:54:05.548284337Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:54:20.32114048Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":32,"history_lines":1,"events_offset":65,"events_lines":2,"console_offset":60,"console_lines":1}
+{"time":"2026-08-14T21:54:20.45358234Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:54:35.321129351Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":33,"history_lines":1,"events_offset":67,"events_lines":2,"console_offset":60,"console_lines":1}
+{"time":"2026-08-14T21:54:35.474897162Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:54:50.321706148Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":34,"history_lines":1,"events_offset":69,"events_lines":2,"console_offset":60,"console_lines":1}
+{"time":"2026-08-14T21:54:50.608442776Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:55:05.321443012Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":71,"events_lines":2,"console_offset":62,"console_lines":9}
+{"time":"2026-08-14T21:55:05.477345408Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:55:20.320811721Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":35,"history_lines":1,"events_offset":73,"events_lines":2,"console_offset":70,"console_lines":2}
+{"time":"2026-08-14T21:55:20.499424268Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:55:35.321629144Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":36,"history_lines":1,"events_offset":75,"events_lines":2,"console_offset":70,"console_lines":1}
+{"time":"2026-08-14T21:55:35.445775297Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:55:50.321311027Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":37,"history_lines":1,"events_offset":77,"events_lines":2,"console_offset":70,"console_lines":1}
+{"time":"2026-08-14T21:55:50.431282685Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:56:05.320926776Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":38,"history_lines":1,"events_offset":79,"events_lines":2,"console_offset":70,"console_lines":1}
+{"time":"2026-08-14T21:56:05.517090556Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:56:20.321730866Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":39,"history_lines":1,"events_offset":81,"events_lines":2,"console_offset":70,"console_lines":1}
+{"time":"2026-08-14T21:56:20.570971062Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:56:35.321155047Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":40,"history_lines":1,"events_offset":83,"events_lines":2,"console_offset":72,"console_lines":10}
+{"time":"2026-08-14T21:56:35.437882867Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:56:50.32102971Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":41,"history_lines":1,"events_offset":85,"events_lines":2,"console_offset":80,"console_lines":1}
+{"time":"2026-08-14T21:56:50.524971529Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:57:05.321204634Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":42,"history_lines":1,"events_offset":87,"events_lines":2,"console_offset":80,"console_lines":1}
+{"time":"2026-08-14T21:57:05.531275159Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:57:20.32172013Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":43,"history_lines":1,"events_offset":89,"events_lines":2,"console_offset":80,"console_lines":1}
+{"time":"2026-08-14T21:57:20.528511373Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:57:35.321170348Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":44,"history_lines":1,"events_offset":91,"events_lines":2,"console_offset":80,"console_lines":1}
+{"time":"2026-08-14T21:57:35.458549034Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:57:50.320862066Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":45,"history_lines":1,"events_offset":93,"events_lines":2,"console_offset":82,"console_lines":10}
+{"time":"2026-08-14T21:57:50.444730503Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:58:05.321650213Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":46,"history_lines":1,"events_offset":95,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-14T21:58:05.534409376Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:58:20.321042008Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":47,"history_lines":1,"events_offset":97,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-14T21:58:20.442037868Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:58:35.32097911Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":48,"history_lines":1,"events_offset":99,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-14T21:58:35.451754594Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:58:50.321128605Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":49,"history_lines":1,"events_offset":101,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-14T21:58:50.456537571Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:59:01.564627783Z","level":"INFO","msg":"fileTransfer: Close: file transfer manager closed"}
+{"time":"2026-08-14T21:59:01.564836113Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":50,"history_lines":3,"events_offset":103,"events_lines":1,"console_offset":90,"console_lines":1,"uploaded_len":3}
+{"time":"2026-08-14T21:59:01.777581888Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:59:01.777738913Z","level":"INFO","msg":"filestream: sending request","total_files":1,"console_offset":92,"console_lines":18,"complete":true,"exit_code":0}
+{"time":"2026-08-14T21:59:01.912393768Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:59:01.913517802Z","level":"INFO","msg":"handler: operation stats","stats":{}}
+{"time":"2026-08-14T21:59:01.916175371Z","level":"INFO","msg":"stream: finishing up"}
+{"time":"2026-08-14T21:59:01.916187489Z","level":"INFO","msg":"handler: closed"}
+{"time":"2026-08-14T21:59:01.916300031Z","level":"INFO","msg":"sender: closed"}
+{"time":"2026-08-14T21:59:01.916303725Z","level":"INFO","msg":"stream: all finished"}

zain/Activation/wandb/run-20260814_214549-k24y90ws/logs/debug.log

@@ -0,0 +1,24 @@
+2026-08-14 21:45:49,463 INFO    MainThread:1229423 [wandb_init.py:setup_run_log_directory():729] Logging user logs to /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/wandb/run-20260814_214549-k24y90ws/logs/debug.log
+2026-08-14 21:45:49,463 INFO    MainThread:1229423 [wandb_init.py:setup_run_log_directory():730] Logging internal logs to /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/wandb/run-20260814_214549-k24y90ws/logs/debug-internal.log
+2026-08-14 21:45:49,463 INFO    MainThread:1229423 [wandb_init.py:init():772] calling init triggers
+2026-08-14 21:45:49,463 INFO    MainThread:1229423 [wandb_init.py:init():777] wandb.init called with sweep_config: {}
+config: {'_wandb': {}}
+2026-08-14 21:45:49,463 INFO    MainThread:1229423 [wandb_init.py:init():820] starting backend
+2026-08-14 21:45:49,463 INFO    MainThread:1229423 [wandb_init.py:init():835] sending inform_init request
+2026-08-14 21:45:49,734 INFO    MainThread:1229423 [wandb_init.py:init():840] backend started and connected
+2026-08-14 21:45:49,735 INFO    MainThread:1229423 [wandb_init.py:init():910] updated telemetry
+2026-08-14 21:45:49,742 INFO    MainThread:1229423 [wandb_init.py:init():933] communicating run to backend with 90.0 second timeout
+2026-08-14 21:45:50,240 INFO    MainThread:1229423 [wandb_init.py:init():978] starting run threads in backend
+2026-08-14 21:45:50,315 INFO    MainThread:1229423 [wandb_run.py:_console_start():2621] atexit reg
+2026-08-14 21:45:50,315 INFO    MainThread:1229423 [wandb_run.py:_redirect():2471] redirect: wrap_raw
+2026-08-14 21:45:50,315 INFO    MainThread:1229423 [wandb_run.py:_redirect():2540] Wrapping output streams.
+2026-08-14 21:45:50,316 INFO    MainThread:1229423 [wandb_run.py:_redirect():2563] Redirects installed.
+2026-08-14 21:45:50,316 INFO    MainThread:1229423 [wandb_init.py:init():1016] run started, returning control to user process
+2026-08-14 21:45:50,317 INFO    MainThread:1229423 [wandb_run.py:_config_callback():1346] config_cb None None {'transformers_version': '5.16.0.dev0', 'architectures': None, 'output_hidden_states': False, 'return_dict': True, 'dtype': None, 'chunk_size_feed_forward': 0, 'is_encoder_decoder': False, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'problem_type': None, 'vocab_size': 4096, 'hidden_size': 128, 'intermediate_size': 256, 'num_hidden_layers': 21, 'num_attention_heads': 4, 'num_key_value_heads': 4, 'hidden_act': 'silu', 'max_position_embeddings': 512, 'initializer_range': 0.02, 'rms_norm_eps': 1e-06, 'use_cache': False, 'pad_token_id': 0, 'bos_token_id': 1, 'eos_token_id': 2, 'pretraining_tp': 1, 'tie_word_embeddings': True, 'rope_parameters': {'rope_theta': 10000.0, 'rope_type': 'default'}, 'attention_bias': False, 'attention_dropout': 0.0, 'mlp_bias': False, 'head_dim': 32, '_name_or_path': '', 'tokenizer_name': 'w-ahmad/tiny-stories-tokenizer', 'mlp_type': 'glu', 'activation': 'silu-waleed10', 'waleed_beta': 10.0, 'model_type': 'tiny_llama', 'output_attentions': False, 'output_dir': 'out/glu-silu-waleed10-21L_run', 'per_device_train_batch_size': 128, 'num_train_epochs': 1, 'max_steps': 1000, 'learning_rate': 0.001, 'lr_scheduler_type': 'constant', 'lr_scheduler_kwargs': None, 'warmup_steps': 0, 'optim': 'adamw_torch', 'optim_args': None, 'weight_decay': 0.01, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'optim_target_modules': None, 'gradient_accumulation_steps': 4, 'average_tokens_across_devices': True, 'max_grad_norm': 1.0, 'label_smoothing_factor': 0.0, 'bf16': True, 'fp16': False, 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'gradient_checkpointing': False, 'gradient_checkpointing_kwargs': None, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'use_liger_kernel': False, 'liger_kernel_config': None, 'neftune_noise_alpha': None, 'torch_empty_cache_steps': None, 'auto_find_batch_size': False, 'logging_strategy': 'steps', 'logging_steps': 20, 'logging_first_step': False, 'log_on_each_node': True, 'logging_nan_inf_filter': True, 'include_num_input_tokens_seen': 'no', 'log_level': 'passive', 'log_level_replica': 'warning', 'disable_tqdm': False, 'report_to': ['wandb'], 'run_name': 'LM-glu-silu-waleed10-21L-4.0M-20260814-214549', 'project': 'huggingface', 'trackio_space_id': None, 'trackio_bucket_id': None, 'trackio_static_space_id': None, 'eval_strategy': 'steps', 'eval_steps': 60000, 'eval_delay': 0, 'per_device_eval_batch_size': 128, 'prediction_loss_only': False, 'eval_on_start': False, 'eval_do_concat_batches': True, 'eval_use_gather_object': False, 'eval_accumulation_steps': None, 'include_for_metrics': [], 'batch_eval_metrics': False, 'save_only_model': False, 'save_strategy': 'steps', 'save_steps': 100, 'save_on_each_node': False, 'save_total_limit': None, 'enable_jit_checkpoint': False, 'push_to_hub': False, 'hub_token': '<HUB_TOKEN>', 'hub_private_repo': None, 'hub_model_id': 'w-ahmad/A-glu-silu-waleed10-21L', 'hub_strategy': 'every_save', 'hub_always_push': False, 'hub_revision': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'restore_callback_states_from_checkpoint': False, 'full_determinism': False, 'seed': 42, 'data_seed': 42, 'use_cpu': False, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}, 'parallelism_config': None, 'dataloader_drop_last': False, 'dataloader_num_workers': 0, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'dataloader_prefetch_factor': None, 'dataloader_multiprocessing_context': None, 'dataloader_in_order': True, 'remove_unused_columns': False, 'label_names': None, 'train_sampling_strategy': 'random', 'length_column_name': 'length', 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'ddp_static_graph': None, 'ddp_backend': None, 'ddp_timeout': 1800, 'fsdp': None, 'fsdp_config': None, 'deepspeed': None, 'debug': [], 'skip_memory_metrics': True, 'do_train': False, 'do_eval': True, 'do_predict': False, 'resume_from_checkpoint': None, 'local_rank': -1}
+2026-08-14 21:45:50,319 INFO    MainThread:1229423 [wandb_config.py:__setitem__():155] [no run ID] config set model/num_parameters = 3970432 - <bound method Run._config_callback of <wandb.sdk.wandb_run.Run object at 0x1528e0152790>>
+2026-08-14 21:45:50,319 INFO    MainThread:1229423 [wandb_run.py:_config_callback():1346] config_cb model/num_parameters 3970432 None
+2026-08-14 21:59:01,157 INFO    MainThread:1229423 [wandb_run.py:_finish():2383] finishing run deepnevro-deepnevro/research1/k24y90ws
+2026-08-14 21:59:01,158 INFO    MainThread:1229423 [wandb_run.py:_atexit_cleanup():2588] got exitcode: 0
+2026-08-14 21:59:01,158 INFO    MainThread:1229423 [wandb_run.py:_restore():2570] restore
+2026-08-14 21:59:01,158 INFO    MainThread:1229423 [wandb_run.py:_restore():2576] restore done
+2026-08-14 21:59:01,915 INFO    MainThread:1229423 [wandb_run.py:_footer_sync_info():3993] logging synced files

zain/Activation/wandb/run-20260814_214549-k24y90ws/run-k24y90ws.wandb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:848becffaac977b2c893e1398902ee9ad2c0cf5857f01232497e0c98c9682d39
+size 563608

zain/Activation/wandb/run-20260814_215902-npap5k3g/files/config.yaml

@@ -0,0 +1,439 @@
+_name_or_path:
+    value: ""
+_wandb:
+    value:
+        cli_version: 0.28.1
+        e:
+            uyw2t8oqcydlnz3g1p7dvon8gcx3um7p:
+                args:
+                    - --config
+                    - /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/configs/baseline.yaml
+                    - --variants
+                    - glu-waleed10
+                    - glu-silu-waleed10
+                    - mlp-waleed10
+                    - mlp-silu-waleed10
+                    - glu-waleed
+                    - glu-situglu_low
+                    - glu-waleedglu_low
+                codePath: sweep.py
+                codePathLocal: sweep.py
+                cpu_count: 112
+                cpu_count_logical: 224
+                cudaVersion: "12.4"
+                disk:
+                    /:
+                        total: "1560765693952"
+                        used: "709083283456"
+                email: deepnevro@gmail.com
+                executable: /mnt/data/zainulabideen/zain-exp/notebooks/my_env/bin/python
+                git:
+                    commit: 3fb47c6943d6927eafc39bd399a3d3a520bae74a
+                    remote: https://github.com/w-ahmad1a10/Activation.git
+                gpu: NVIDIA H100 80GB HBM3
+                gpu_count: 8
+                gpu_nvidia:
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-39c684a5-fde6-83d7-1663-0859795881ae
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-68012e5a-38b6-b643-0ca6-62fb66720bf3
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-132944c4-b689-2b5f-89a4-d730401677ab
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-2df386cc-6d26-d0e2-7a2d-a057b0d95864
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-bfa16575-1d94-1aa2-4537-2c93433f42ef
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-bc6c3e3c-9b90-09ca-c034-774961847c54
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-00a441e1-7c95-e7d6-4c35-43d6b291aea9
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-1c4d29a2-4647-6fce-d8fc-0c5ecfbbd6ea
+                host: deeplens-k3s-node1
+                memory:
+                    total: "2164089937920"
+                os: Linux-5.15.0-126-generic-x86_64-with-glibc2.35
+                program: /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/sweep.py
+                python: CPython 3.11.15
+                root: /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation
+                startedAt: "2026-08-14T21:59:02.522762Z"
+                writerId: uyw2t8oqcydlnz3g1p7dvon8gcx3um7p
+        m:
+            - "1": train/global_step
+              "6":
+                - 3
+              "7": []
+            - "2": '*'
+              "5": 1
+              "6":
+                - 1
+              "7": []
+        python_version: 3.11.15
+        t:
+            "1":
+                - 1
+                - 5
+                - 11
+                - 41
+                - 49
+                - 51
+                - 53
+                - 71
+            "2":
+                - 1
+                - 5
+                - 11
+                - 41
+                - 49
+                - 51
+                - 53
+                - 71
+            "3":
+                - 2
+                - 7
+                - 13
+                - 19
+                - 62
+                - 66
+            "4": 3.11.15
+            "5": 0.28.1
+            "6": 5.16.0.dev0
+            "9":
+                "1": transformers_trainer
+            "12": 0.28.1
+            "13": linux-x86_64
+accelerator_config:
+    value:
+        dispatch_batches: null
+        even_batches: true
+        gradient_accumulation_kwargs: null
+        non_blocking: false
+        split_batches: false
+        use_seedable_sampler: true
+activation:
+    value: waleed10
+adam_beta1:
+    value: 0.9
+adam_beta2:
+    value: 0.999
+adam_epsilon:
+    value: 1e-08
+architectures:
+    value: null
+attention_bias:
+    value: false
+attention_dropout:
+    value: 0
+auto_find_batch_size:
+    value: false
+average_tokens_across_devices:
+    value: true
+batch_eval_metrics:
+    value: false
+bf16:
+    value: true
+bf16_full_eval:
+    value: false
+bos_token_id:
+    value: 1
+chunk_size_feed_forward:
+    value: 0
+data_seed:
+    value: 42
+dataloader_drop_last:
+    value: false
+dataloader_in_order:
+    value: true
+dataloader_multiprocessing_context:
+    value: null
+dataloader_num_workers:
+    value: 0
+dataloader_persistent_workers:
+    value: false
+dataloader_pin_memory:
+    value: true
+dataloader_prefetch_factor:
+    value: null
+ddp_backend:
+    value: null
+ddp_broadcast_buffers:
+    value: null
+ddp_bucket_cap_mb:
+    value: null
+ddp_find_unused_parameters:
+    value: null
+ddp_static_graph:
+    value: null
+ddp_timeout:
+    value: 1800
+debug:
+    value: []
+deepspeed:
+    value: null
+disable_tqdm:
+    value: false
+do_eval:
+    value: true
+do_predict:
+    value: false
+do_train:
+    value: false
+dtype:
+    value: null
+enable_jit_checkpoint:
+    value: false
+eos_token_id:
+    value: 2
+eval_accumulation_steps:
+    value: null
+eval_delay:
+    value: 0
+eval_do_concat_batches:
+    value: true
+eval_on_start:
+    value: false
+eval_steps:
+    value: 60000
+eval_strategy:
+    value: steps
+eval_use_gather_object:
+    value: false
+fp16:
+    value: false
+fp16_full_eval:
+    value: false
+fsdp:
+    value: null
+fsdp_config:
+    value: null
+full_determinism:
+    value: false
+gradient_accumulation_steps:
+    value: 4
+gradient_checkpointing:
+    value: false
+gradient_checkpointing_kwargs:
+    value: null
+greater_is_better:
+    value: null
+head_dim:
+    value: 32
+hidden_act:
+    value: silu
+hidden_size:
+    value: 128
+hub_always_push:
+    value: false
+hub_model_id:
+    value: w-ahmad/A-mlp-waleed10-21L
+hub_private_repo:
+    value: null
+hub_revision:
+    value: null
+hub_strategy:
+    value: every_save
+hub_token:
+    value: <HUB_TOKEN>
+id2label:
+    value:
+        "0": LABEL_0
+        "1": LABEL_1
+ignore_data_skip:
+    value: false
+include_for_metrics:
+    value: []
+include_num_input_tokens_seen:
+    value: "no"
+initializer_range:
+    value: 0.02
+intermediate_size:
+    value: 256
+is_encoder_decoder:
+    value: false
+label_names:
+    value: null
+label_smoothing_factor:
+    value: 0
+label2id:
+    value:
+        LABEL_0: 0
+        LABEL_1: 1
+learning_rate:
+    value: 0.001
+length_column_name:
+    value: length
+liger_kernel_config:
+    value: null
+load_best_model_at_end:
+    value: false
+local_rank:
+    value: -1
+log_level:
+    value: passive
+log_level_replica:
+    value: warning
+log_on_each_node:
+    value: true
+logging_first_step:
+    value: false
+logging_nan_inf_filter:
+    value: true
+logging_steps:
+    value: 20
+logging_strategy:
+    value: steps
+lr_scheduler_kwargs:
+    value: null
+lr_scheduler_type:
+    value: constant
+max_grad_norm:
+    value: 1
+max_position_embeddings:
+    value: 512
+max_steps:
+    value: 1000
+metric_for_best_model:
+    value: null
+mlp_bias:
+    value: false
+mlp_type:
+    value: mlp
+model/num_parameters:
+    value: 3970432
+model_type:
+    value: tiny_llama
+neftune_noise_alpha:
+    value: null
+num_attention_heads:
+    value: 4
+num_hidden_layers:
+    value: 21
+num_key_value_heads:
+    value: 4
+num_train_epochs:
+    value: 1
+optim:
+    value: adamw_torch
+optim_args:
+    value: null
+optim_target_modules:
+    value: null
+output_attentions:
+    value: false
+output_dir:
+    value: out/mlp-waleed10-21L_run
+output_hidden_states:
+    value: false
+pad_token_id:
+    value: 0
+parallelism_config:
+    value: null
+per_device_eval_batch_size:
+    value: 128
+per_device_train_batch_size:
+    value: 128
+prediction_loss_only:
+    value: false
+pretraining_tp:
+    value: 1
+problem_type:
+    value: null
+project:
+    value: huggingface
+push_to_hub:
+    value: false
+remove_unused_columns:
+    value: false
+report_to:
+    value:
+        - wandb
+restore_callback_states_from_checkpoint:
+    value: false
+resume_from_checkpoint:
+    value: null
+return_dict:
+    value: true
+rms_norm_eps:
+    value: 1e-06
+rope_parameters:
+    value:
+        rope_theta: 10000
+        rope_type: default
+run_name:
+    value: LM-mlp-waleed10-21L-4.0M-20260814-215902
+save_on_each_node:
+    value: false
+save_only_model:
+    value: false
+save_steps:
+    value: 100
+save_strategy:
+    value: steps
+save_total_limit:
+    value: null
+seed:
+    value: 42
+skip_memory_metrics:
+    value: true
+tf32:
+    value: null
+tie_word_embeddings:
+    value: true
+tokenizer_name:
+    value: w-ahmad/tiny-stories-tokenizer
+torch_compile:
+    value: false
+torch_compile_backend:
+    value: null
+torch_compile_mode:
+    value: null
+torch_empty_cache_steps:
+    value: null
+trackio_bucket_id:
+    value: null
+trackio_space_id:
+    value: null
+trackio_static_space_id:
+    value: null
+train_sampling_strategy:
+    value: random
+transformers_version:
+    value: 5.16.0.dev0
+use_cache:
+    value: false
+use_cpu:
+    value: false
+use_liger_kernel:
+    value: false
+vocab_size:
+    value: 4096
+waleed_beta:
+    value: 10
+warmup_steps:
+    value: 0
+weight_decay:
+    value: 0.01

zain/Activation/wandb/run-20260814_215902-npap5k3g/files/output.log

@@ -0,0 +1,110 @@
+ 10%|█         | 100/1000 [01:15<11:13,  1.34it/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': '29.1', 'grad_norm': '4.219', 'learning_rate': '0.001', 'epoch': '0.01078', 'train/total_time_seconds': '7.459', 'train/time_per_step_avg': '0.3729', 'train/epoch_time_elapsed': '15.17', 'train/estimated_remaining_minutes': '6.091'}
+{'loss': '24.02', 'grad_norm': '13.25', 'learning_rate': '0.001', 'epoch': '0.02157', 'train/total_time_seconds': '14.82', 'train/time_per_step_avg': '0.3705', 'train/epoch_time_elapsed': '30.08', 'train/estimated_remaining_minutes': '5.928'}
+{'loss': '22.07', 'grad_norm': '2.156', 'learning_rate': '0.001', 'epoch': '0.03235', 'train/total_time_seconds': '22.18', 'train/time_per_step_avg': '0.3696', 'train/epoch_time_elapsed': '45.2', 'train/estimated_remaining_minutes': '5.79'}
+{'loss': '20.73', 'grad_norm': '2.984', 'learning_rate': '0.001', 'epoch': '0.04314', 'train/total_time_seconds': '29.57', 'train/time_per_step_avg': '0.3696', 'train/epoch_time_elapsed': '60.31', 'train/estimated_remaining_minutes': '5.668'}
+{'loss': '19.62', 'grad_norm': '2.562', 'learning_rate': '0.001', 'epoch': '0.05392', 'train/total_time_seconds': '36.97', 'train/time_per_step_avg': '0.3697', 'train/epoch_time_elapsed': '75.41', 'train/estimated_remaining_minutes': '5.545'}
+  - 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, 43.34it/s]
+ 20%|██        | 200/1000 [02:30<09:53,  1.35it/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': '18.49', 'grad_norm': '3.5', 'learning_rate': '0.001', 'epoch': '0.06471', 'train/total_time_seconds': '44.36', 'train/time_per_step_avg': '0.369', 'train/epoch_time_elapsed': '90.44', 'train/estimated_remaining_minutes': '5.422'}
+{'loss': '17.53', 'grad_norm': '2.328', 'learning_rate': '0.001', 'epoch': '0.07549', 'train/total_time_seconds': '51.76', 'train/time_per_step_avg': '0.3694', 'train/epoch_time_elapsed': '105.6', 'train/estimated_remaining_minutes': '5.299'}
+{'loss': '16.74', 'grad_norm': '4', 'learning_rate': '0.001', 'epoch': '0.08628', 'train/total_time_seconds': '59.16', 'train/time_per_step_avg': '0.3698', 'train/epoch_time_elapsed': '120.5', 'train/estimated_remaining_minutes': '5.176'}
+{'loss': '16.19', 'grad_norm': '2.656', 'learning_rate': '0.001', 'epoch': '0.09706', 'train/total_time_seconds': '66.55', 'train/time_per_step_avg': '0.3698', 'train/epoch_time_elapsed': '135.5', 'train/estimated_remaining_minutes': '5.053'}
+{'loss': '15.68', 'grad_norm': '2.047', 'learning_rate': '0.001', 'epoch': '0.1078', 'train/total_time_seconds': '73.92', 'train/time_per_step_avg': '0.3696', 'train/epoch_time_elapsed': '150.5', 'train/estimated_remaining_minutes': '4.928'}
+  - 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, 78.32it/s]
+ 30%|███       | 300/1000 [03:45<08:42,  1.34it/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': '15.22', 'grad_norm': '3.906', 'learning_rate': '0.001', 'epoch': '0.1186', 'train/total_time_seconds': '81.28', 'train/time_per_step_avg': '0.3691', 'train/epoch_time_elapsed': '165.5', 'train/estimated_remaining_minutes': '4.803'}
+{'loss': '14.92', 'grad_norm': '2.75', 'learning_rate': '0.001', 'epoch': '0.1294', 'train/total_time_seconds': '88.63', 'train/time_per_step_avg': '0.3687', 'train/epoch_time_elapsed': '180.4', 'train/estimated_remaining_minutes': '4.678'}
+{'loss': '14.54', 'grad_norm': '1.633', 'learning_rate': '0.001', 'epoch': '0.1402', 'train/total_time_seconds': '96', 'train/time_per_step_avg': '0.3684', 'train/epoch_time_elapsed': '195.6', 'train/estimated_remaining_minutes': '4.554'}
+{'loss': '14.27', 'grad_norm': '2.688', 'learning_rate': '0.001', 'epoch': '0.151', 'train/total_time_seconds': '103.4', 'train/time_per_step_avg': '0.3684', 'train/epoch_time_elapsed': '210.5', 'train/estimated_remaining_minutes': '4.431'}
+{'loss': '14.02', 'grad_norm': '3.156', 'learning_rate': '0.001', 'epoch': '0.1618', 'train/total_time_seconds': '110.8', 'train/time_per_step_avg': '0.3687', 'train/epoch_time_elapsed': '225.5', 'train/estimated_remaining_minutes': '4.308'}
+  - 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, 80.87it/s]
+ 40%|████      | 400/1000 [05:00<07:42,  1.30it/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': '13.78', 'grad_norm': '3.938', 'learning_rate': '0.001', 'epoch': '0.1726', 'train/total_time_seconds': '118.2', 'train/time_per_step_avg': '0.369', 'train/epoch_time_elapsed': '240.5', 'train/estimated_remaining_minutes': '4.186'}
+{'loss': '13.6', 'grad_norm': '2.906', 'learning_rate': '0.001', 'epoch': '0.1833', 'train/total_time_seconds': '125.6', 'train/time_per_step_avg': '0.3695', 'train/epoch_time_elapsed': '255.5', 'train/estimated_remaining_minutes': '4.063'}
+{'loss': '13.37', 'grad_norm': '2.219', 'learning_rate': '0.001', 'epoch': '0.1941', 'train/total_time_seconds': '133', 'train/time_per_step_avg': '0.3697', 'train/epoch_time_elapsed': '270.5', 'train/estimated_remaining_minutes': '3.94'}
+{'loss': '13.15', 'grad_norm': '3.797', 'learning_rate': '0.001', 'epoch': '0.2049', 'train/total_time_seconds': '140.3', 'train/time_per_step_avg': '0.3693', 'train/epoch_time_elapsed': '285.5', 'train/estimated_remaining_minutes': '3.816'}
+{'loss': '13.03', 'grad_norm': '1.859', 'learning_rate': '0.001', 'epoch': '0.2157', 'train/total_time_seconds': '147.7', 'train/time_per_step_avg': '0.3689', 'train/epoch_time_elapsed': '300.6', 'train/estimated_remaining_minutes': '3.692'}
+  - 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, 83.27it/s]
+ 50%|█████     | 500/1000 [06:15<06:15,  1.33it/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': '12.84', 'grad_norm': '2.766', 'learning_rate': '0.001', 'epoch': '0.2265', 'train/total_time_seconds': '155', 'train/time_per_step_avg': '0.3687', 'train/epoch_time_elapsed': '315.7', 'train/estimated_remaining_minutes': '3.569'}
+{'loss': '12.73', 'grad_norm': '4.531', 'learning_rate': '0.001', 'epoch': '0.2373', 'train/total_time_seconds': '162.4', 'train/time_per_step_avg': '0.3687', 'train/epoch_time_elapsed': '330.6', 'train/estimated_remaining_minutes': '3.446'}
+{'loss': '12.55', 'grad_norm': '3.031', 'learning_rate': '0.001', 'epoch': '0.248', 'train/total_time_seconds': '169.8', 'train/time_per_step_avg': '0.3688', 'train/epoch_time_elapsed': '345.7', 'train/estimated_remaining_minutes': '3.323'}
+{'loss': '12.41', 'grad_norm': '3.359', 'learning_rate': '0.001', 'epoch': '0.2588', 'train/total_time_seconds': '177.3', 'train/time_per_step_avg': '0.3693', 'train/epoch_time_elapsed': '360.6', 'train/estimated_remaining_minutes': '3.2'}
+{'loss': '12.28', 'grad_norm': '3.609', 'learning_rate': '0.001', 'epoch': '0.2696', 'train/total_time_seconds': '184.7', 'train/time_per_step_avg': '0.3698', 'train/epoch_time_elapsed': '375.6', 'train/estimated_remaining_minutes': '3.078'}
+  - 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, 66.41it/s]
+ 60%|██████    | 600/1000 [07:30<04:57,  1.34it/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': '12.15', 'grad_norm': '2.406', 'learning_rate': '0.001', 'epoch': '0.2804', 'train/total_time_seconds': '192', 'train/time_per_step_avg': '0.37', 'train/epoch_time_elapsed': '390.7', 'train/estimated_remaining_minutes': '2.955'}
+{'loss': '12.05', 'grad_norm': '3', 'learning_rate': '0.001', 'epoch': '0.2912', 'train/total_time_seconds': '199.4', 'train/time_per_step_avg': '0.37', 'train/epoch_time_elapsed': '405.9', 'train/estimated_remaining_minutes': '2.832'}
+{'loss': '11.96', 'grad_norm': '2.547', 'learning_rate': '0.001', 'epoch': '0.302', 'train/total_time_seconds': '206.8', 'train/time_per_step_avg': '0.3698', 'train/epoch_time_elapsed': '421', 'train/estimated_remaining_minutes': '2.709'}
+{'loss': '11.82', 'grad_norm': '2.25', 'learning_rate': '0.001', 'epoch': '0.3128', 'train/total_time_seconds': '214.2', 'train/time_per_step_avg': '0.3697', 'train/epoch_time_elapsed': '436', 'train/estimated_remaining_minutes': '2.585'}
+{'loss': '11.72', 'grad_norm': '3.219', 'learning_rate': '0.001', 'epoch': '0.3235', 'train/total_time_seconds': '221.6', 'train/time_per_step_avg': '0.3694', 'train/epoch_time_elapsed': '450.9', 'train/estimated_remaining_minutes': '2.462'}
+  - 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, 86.34it/s]
+ 70%|███████   | 700/1000 [08:45<03:42,  1.35it/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': '11.65', 'grad_norm': '2.438', 'learning_rate': '0.001', 'epoch': '0.3343', 'train/total_time_seconds': '229', 'train/time_per_step_avg': '0.3692', 'train/epoch_time_elapsed': '465.9', 'train/estimated_remaining_minutes': '2.339'}
+{'loss': '11.56', 'grad_norm': '2.469', 'learning_rate': '0.001', 'epoch': '0.3451', 'train/total_time_seconds': '236.4', 'train/time_per_step_avg': '0.369', 'train/epoch_time_elapsed': '480.9', 'train/estimated_remaining_minutes': '2.216'}
+{'loss': '11.47', 'grad_norm': '2.5', 'learning_rate': '0.001', 'epoch': '0.3559', 'train/total_time_seconds': '243.7', 'train/time_per_step_avg': '0.3688', 'train/epoch_time_elapsed': '495.9', 'train/estimated_remaining_minutes': '2.093'}
+{'loss': '11.37', 'grad_norm': '2.906', 'learning_rate': '0.001', 'epoch': '0.3667', 'train/total_time_seconds': '251.1', 'train/time_per_step_avg': '0.3689', 'train/epoch_time_elapsed': '511.1', 'train/estimated_remaining_minutes': '1.969'}
+{'loss': '11.29', 'grad_norm': '3.141', 'learning_rate': '0.001', 'epoch': '0.3775', 'train/total_time_seconds': '258.5', 'train/time_per_step_avg': '0.3689', 'train/epoch_time_elapsed': '526', 'train/estimated_remaining_minutes': '1.846'}
+  - 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, 86.72it/s]
+ 80%|████████  | 800/1000 [10:01<02:29,  1.34it/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': '11.2', 'grad_norm': '2.969', 'learning_rate': '0.001', 'epoch': '0.3882', 'train/total_time_seconds': '265.9', 'train/time_per_step_avg': '0.369', 'train/epoch_time_elapsed': '541.2', 'train/estimated_remaining_minutes': '1.723'}
+{'loss': '11.14', 'grad_norm': '2.734', 'learning_rate': '0.001', 'epoch': '0.399', 'train/total_time_seconds': '273.3', 'train/time_per_step_avg': '0.369', 'train/epoch_time_elapsed': '556.1', 'train/estimated_remaining_minutes': '1.6'}
+{'loss': '11.09', 'grad_norm': '2.219', 'learning_rate': '0.001', 'epoch': '0.4098', 'train/total_time_seconds': '280.6', 'train/time_per_step_avg': '0.3692', 'train/epoch_time_elapsed': '571.3', 'train/estimated_remaining_minutes': '1.477'}
+{'loss': '11.03', 'grad_norm': '2.797', 'learning_rate': '0.001', 'epoch': '0.4206', 'train/total_time_seconds': '288', 'train/time_per_step_avg': '0.3692', 'train/epoch_time_elapsed': '586.2', 'train/estimated_remaining_minutes': '1.354'}
+{'loss': '10.93', 'grad_norm': '3.359', 'learning_rate': '0.001', 'epoch': '0.4314', 'train/total_time_seconds': '295.4', 'train/time_per_step_avg': '0.3693', 'train/epoch_time_elapsed': '601.1', 'train/estimated_remaining_minutes': '1.231'}
+  - 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, 83.92it/s]
+ 90%|█████████ | 900/1000 [11:16<01:14,  1.33it/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': '10.88', 'grad_norm': '2.469', 'learning_rate': '0.001', 'epoch': '0.4422', 'train/total_time_seconds': '302.8', 'train/time_per_step_avg': '0.3692', 'train/epoch_time_elapsed': '616.1', 'train/estimated_remaining_minutes': '1.108'}
+{'loss': '10.83', 'grad_norm': '2.797', 'learning_rate': '0.001', 'epoch': '0.453', 'train/total_time_seconds': '310.2', 'train/time_per_step_avg': '0.3691', 'train/epoch_time_elapsed': '631.3', 'train/estimated_remaining_minutes': '0.9847'}
+{'loss': '10.75', 'grad_norm': '2.469', 'learning_rate': '0.001', 'epoch': '0.4637', 'train/total_time_seconds': '317.5', 'train/time_per_step_avg': '0.369', 'train/epoch_time_elapsed': '646.3', 'train/estimated_remaining_minutes': '0.8615'}
+{'loss': '10.71', 'grad_norm': '2.844', 'learning_rate': '0.001', 'epoch': '0.4745', 'train/total_time_seconds': '324.9', 'train/time_per_step_avg': '0.369', 'train/epoch_time_elapsed': '661.4', 'train/estimated_remaining_minutes': '0.7385'}
+{'loss': '10.66', 'grad_norm': '2.75', 'learning_rate': '0.001', 'epoch': '0.4853', 'train/total_time_seconds': '332.3', 'train/time_per_step_avg': '0.3687', 'train/epoch_time_elapsed': '676.4', 'train/estimated_remaining_minutes': '0.6153'}
+  - 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, 85.12it/s]
+100%|██████████| 1000/1000 [12:41<00:00,  1.31i[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': '10.58', 'grad_norm': '2.422', 'learning_rate': '0.001', 'epoch': '0.4961', 'train/total_time_seconds': '339.7', 'train/time_per_step_avg': '0.3687', 'train/epoch_time_elapsed': '691.7', 'train/estimated_remaining_minutes': '0.4923'}
+{'loss': '10.56', 'grad_norm': '2.219', 'learning_rate': '0.001', 'epoch': '0.5069', 'train/total_time_seconds': '347.1', 'train/time_per_step_avg': '0.369', 'train/epoch_time_elapsed': '706.7', 'train/estimated_remaining_minutes': '0.3692'}
+{'loss': '10.53', 'grad_norm': '2.328', 'learning_rate': '0.001', 'epoch': '0.5177', 'train/total_time_seconds': '354.5', 'train/time_per_step_avg': '0.3697', 'train/epoch_time_elapsed': '721.8', 'train/estimated_remaining_minutes': '0.2462'}
+{'loss': '10.48', 'grad_norm': '3.453', 'learning_rate': '0.001', 'epoch': '0.5284', 'train/total_time_seconds': '361.9', 'train/time_per_step_avg': '0.3697', 'train/epoch_time_elapsed': '736.7', 'train/estimated_remaining_minutes': '0.1231'}
+{'loss': '10.44', 'grad_norm': '2.375', 'learning_rate': '0.001', 'epoch': '0.5392', 'train/total_time_seconds': '369.3', 'train/time_per_step_avg': '0.37', 'train/epoch_time_elapsed': '751.8', 'train/estimated_remaining_minutes': '0'}
+  - 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': '2.601', 'eval_runtime': '9.308', 'eval_samples_per_second': '1024', 'eval_steps_per_second': '8.058', 'epoch': '0.5392', 'train/total_time_seconds': '369.3', 'train/time_per_step_avg': '0.37', 'train/epoch_time_elapsed': '761.1', 'train/estimated_remaining_minutes': '0'}
+  - 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, 72.91it/s]
+100%|██████████| 1000/1000 [12:41<00:00,  1.31it/s], ?it/s]
+{'train_runtime': '761.9', 'train_samples_per_second': '672', 'train_steps_per_second': '1.312', 'train_loss': '13.63', 'epoch': '0.5392', 'train/total_time_seconds': '369.3', 'train/time_per_step_avg': '0.37', 'train/epoch_time_elapsed': '761.2', 'train/estimated_remaining_minutes': '0'}
+100%|██████████| 75/75 [00:09<00:00,  8.30it/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.
+  - 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, 88.27it/s]
+>>> FINISHED mlp-waleed10-21L successfully

zain/Activation/wandb/run-20260814_215902-npap5k3g/files/requirements.txt

@@ -0,0 +1,149 @@
+asttokens==3.0.1
+comm==0.2.3
+debugpy==1.8.21
+decorator==5.3.1
+executing==2.2.1
+nest-asyncio==1.6.0
+parso==0.8.7
+platformdirs==4.11.0
+psutil==7.2.2
+ptyprocess==0.7.0
+pure_eval==0.2.3
+Pygments==2.20.0
+pyzmq==27.1.0
+setuptools==83.0.0
+six==1.17.0
+tornado==6.5.7
+traitlets==5.15.0
+fsspec==2026.4.0
+wcwidth==0.8.2
+ipython_pygments_lexers==1.1.1
+jedi==0.20.0
+jupyter_core==5.9.1
+matplotlib-inline==0.2.2
+pexpect==4.9.0
+prompt_toolkit==3.0.53
+python-dateutil==2.9.0.post0
+stack_data==0.6.3
+wheel==0.47.0
+jupyter_client==8.9.1
+pip==26.1.2
+ipython==9.15.0
+ipykernel==7.2.0
+threadpoolctl==3.6.0
+pyparsing==3.3.2
+typing_extensions==4.15.0
+Jinja2==3.1.6
+narwhals==2.24.0
+kiwisolver==1.5.0
+joblib==1.5.3
+fonttools==4.63.0
+cycler==0.12.1
+scipy==1.17.1
+pandas==3.0.5
+contourpy==1.3.3
+scikit-learn==1.9.0
+matplotlib==3.11.1
+urllib3==2.7.0
+tqdm==4.70.0
+idna==3.18
+charset-normalizer==3.4.9
+certifi==2026.7.22
+requests==2.34.2
+seaborn==0.13.2
+uv==0.12.0
+shellingham==1.5.4
+mpmath==1.3.0
+attrs==26.1.0
+hf-xet==1.5.2
+nvidia-nccl-cu12==2.21.5
+MarkupSafe==3.0.3
+regex==2026.7.19
+importlib_metadata==9.0.0
+httpcore==1.0.9
+annotated-doc==0.0.5
+multidict==6.7.1
+aiohttp==3.14.3
+aiosignal==1.4.0
+xxhash==3.8.1
+aiohappyeyeballs==2.7.1
+mdurl==0.1.2
+cuda-toolkit==13.0.3.0
+networkx==3.6.1
+PyYAML==6.0.3
+nvidia-cufile==1.15.1.6
+typer==0.27.0
+torchaudio==2.6.0+cu124
+rich==15.0.0
+nvidia-cufft-cu12==11.2.1.3
+h11==0.16.0
+dill==0.4.1
+cuda-pathfinder==1.6.0
+filelock==3.29.0
+nvidia-nvtx-cu12==12.4.127
+httpx==0.28.1
+anyio==4.14.2
+numpy==2.4.4
+yarl==1.24.5
+click==8.4.2
+triton==3.2.0
+frozenlist==1.8.0
+zipp==4.1.0
+propcache==0.5.2
+tokenizers==0.22.2
+markdown-it-py==4.2.0
+nvidia-cuda-runtime==13.0.96
+cuda-bindings==13.3.1
+nvidia-cuda-cupti==13.0.85
+torch==2.6.0+cu124
+multiprocess==0.70.19
+pillow==12.2.0
+transformers==5.16.0.dev0
+wandb==0.28.1
+nvidia-curand==10.4.0.35
+sympy==1.13.1
+nvidia-cusparse==12.6.3.3
+nvidia-cuda-nvrtc==13.0.88
+typing-inspection==0.4.2
+nvidia-cusolver==12.0.4.66
+nvidia-cufft==12.0.0.61
+nvidia-cudnn-cu13==9.20.0.48
+nvidia-cublas==13.1.1.3
+pyarrow==25.0.0
+evaluate==0.4.6
+diffusers==0.39.0
+pydantic==2.13.4
+annotated-types==0.8.0
+protobuf==7.35.1
+sentry-sdk==2.66.1
+einops==0.8.2
+packaging==26.2
+nvidia-nvjitlink-cu12==12.4.127
+nvidia-curand-cu12==10.3.5.147
+nvidia-cusparselt-cu12==0.6.2
+nvidia-cusparse-cu12==12.3.1.170
+nvidia-cuda-runtime-cu12==12.4.127
+torchvision==0.21.0+cu124
+nvidia-cuda-nvrtc-cu12==12.4.127
+nvidia-cuda-cupti-cu12==12.4.127
+nvidia-cusolver-cu12==11.6.1.9
+nvidia-cublas-cu12==12.4.5.8
+nvidia-cudnn-cu12==9.1.0.70
+huggingface_hub==1.26.0
+datasets==5.0.1
+safetensors==0.8.0
+accelerate==1.14.0
+pydantic_core==2.46.4
+ninja==1.13.0
+autocommand==2.2.2
+backports.tarfile==1.2.0
+importlib_metadata==8.7.1
+jaraco.text==4.0.0
+jaraco.context==6.1.0
+jaraco.functools==4.4.0
+more-itertools==10.8.0
+packaging==26.0
+platformdirs==4.4.0
+tomli==2.4.0
+wheel==0.46.3
+zipp==3.23.0

zain/Activation/wandb/run-20260814_215902-npap5k3g/files/wandb-metadata.json

@@ -0,0 +1,101 @@
+{
+  "os":  "Linux-5.15.0-126-generic-x86_64-with-glibc2.35",
+  "python":  "CPython 3.11.15",
+  "startedAt":  "2026-08-14T21:59:02.522762Z",
+  "args":  [
+    "--config",
+    "/mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/configs/baseline.yaml",
+    "--variants",
+    "glu-waleed10",
+    "glu-silu-waleed10",
+    "mlp-waleed10",
+    "mlp-silu-waleed10",
+    "glu-waleed",
+    "glu-situglu_low",
+    "glu-waleedglu_low"
+  ],
+  "program":  "/mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/sweep.py",
+  "codePath":  "sweep.py",
+  "codePathLocal":  "sweep.py",
+  "git":  {
+    "remote":  "https://github.com/w-ahmad1a10/Activation.git",
+    "commit":  "3fb47c6943d6927eafc39bd399a3d3a520bae74a"
+  },
+  "email":  "deepnevro@gmail.com",
+  "root":  "/mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation",
+  "host":  "deeplens-k3s-node1",
+  "executable":  "/mnt/data/zainulabideen/zain-exp/notebooks/my_env/bin/python",
+  "cpu_count":  112,
+  "cpu_count_logical":  224,
+  "gpu":  "NVIDIA H100 80GB HBM3",
+  "gpu_count":  8,
+  "disk":  {
+    "/":  {
+      "total":  "1560765693952",
+      "used":  "709083283456"
+    }
+  },
+  "memory":  {
+    "total":  "2164089937920"
+  },
+  "gpu_nvidia":  [
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-39c684a5-fde6-83d7-1663-0859795881ae"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-68012e5a-38b6-b643-0ca6-62fb66720bf3"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-132944c4-b689-2b5f-89a4-d730401677ab"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-2df386cc-6d26-d0e2-7a2d-a057b0d95864"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-bfa16575-1d94-1aa2-4537-2c93433f42ef"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-bc6c3e3c-9b90-09ca-c034-774961847c54"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-00a441e1-7c95-e7d6-4c35-43d6b291aea9"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-1c4d29a2-4647-6fce-d8fc-0c5ecfbbd6ea"
+    }
+  ],
+  "cudaVersion":  "12.4",
+  "writerId":  "uyw2t8oqcydlnz3g1p7dvon8gcx3um7p"
+}

zain/Activation/wandb/run-20260814_215902-npap5k3g/files/wandb-summary.json

@@ -0,0 +1 @@
+{"train_samples_per_second":671.984,"_timestamp":1.786745513620422e+09,"eval/steps_per_second":8.125,"train/loss":10.437238311767578,"_step":52,"train/learning_rate":0.001,"total_flos":5.420315836416e+15,"train/global_step":1000,"train/grad_norm":2.375,"_wandb":{"runtime":770},"train_loss":13.634520202636718,"_runtime":770,"train_runtime":761.9233,"eval/samples_per_second":1032.07,"train_steps_per_second":1.312,"eval/loss":2.6014304161071777,"train/epoch":0.5392289026691831,"eval/runtime":9.231}

zain/Activation/wandb/run-20260814_215902-npap5k3g/logs/debug-core.log

@@ -0,0 +1,62 @@
+{"time":"2026-08-14T21:31:53.539152008Z","level":"INFO","msg":"main: starting server","port-filename":"/tmp/tmp650y3sc2/port-1229423.txt","pid":1229423,"detached":false,"idle-timeout":600000000000,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false,"enable-dcgm-profiling":false}
+{"time":"2026-08-14T21:31:53.540676876Z","level":"INFO","msg":"server: will exit if parent process dies","ppid":1229423}
+{"time":"2026-08-14T21:31:53.540659035Z","level":"INFO","msg":"server: accepting connections","addr":{"Name":"/tmp/wandb-1229423-1230766-2604858008/socket","Net":"unix"}}
+{"time":"2026-08-14T21:31:53.717724383Z","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"1(@)"}
+{"time":"2026-08-14T21:31:53.802092564Z","level":"INFO","msg":"handleInformInit: received","streamId":"q9vcmje5","id":"1(@)"}
+{"time":"2026-08-14T21:31:54.062397411Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"q9vcmje5","id":"1(@)"}
+{"time":"2026-08-14T21:31:59.411924652Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"32s1adqk73rr"}
+{"time":"2026-08-14T21:45:48.396635188Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"32s1adqk73rr"}
+{"time":"2026-08-14T21:45:48.929834053Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"32s1adqk73rr"}
+{"time":"2026-08-14T21:45:48.931620539Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"q9vcmje5","id":"1(@)"}
+{"time":"2026-08-14T21:45:48.932491197Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"q9vcmje5","id":"1(@)"}
+{"time":"2026-08-14T21:45:49.464388506Z","level":"INFO","msg":"handleInformInit: received","streamId":"k24y90ws","id":"1(@)"}
+{"time":"2026-08-14T21:45:49.733853364Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"k24y90ws","id":"1(@)"}
+{"time":"2026-08-14T21:45:55.329792921Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"dr322rnt7xgx"}
+{"time":"2026-08-14T21:59:01.158537572Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"dr322rnt7xgx"}
+{"time":"2026-08-14T21:59:01.914569559Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"dr322rnt7xgx"}
+{"time":"2026-08-14T21:59:01.916156477Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"k24y90ws","id":"1(@)"}
+{"time":"2026-08-14T21:59:01.917534874Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"k24y90ws","id":"1(@)"}
+{"time":"2026-08-14T21:59:02.524868682Z","level":"INFO","msg":"handleInformInit: received","streamId":"npap5k3g","id":"1(@)"}
+{"time":"2026-08-14T21:59:02.786199875Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"npap5k3g","id":"1(@)"}
+{"time":"2026-08-14T21:59:08.192858634Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"2g99yy5pussw"}
+{"time":"2026-08-14T22:11:53.643379161Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"2g99yy5pussw"}
+{"time":"2026-08-14T22:11:54.173579992Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"2g99yy5pussw"}
+{"time":"2026-08-14T22:11:54.17490129Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"npap5k3g","id":"1(@)"}
+{"time":"2026-08-14T22:11:54.175479186Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"npap5k3g","id":"1(@)"}
+{"time":"2026-08-14T22:11:54.694802218Z","level":"INFO","msg":"handleInformInit: received","streamId":"a3607q94","id":"1(@)"}
+{"time":"2026-08-14T22:11:54.956642682Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"a3607q94","id":"1(@)"}
+{"time":"2026-08-14T22:12:00.441973218Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"raaqqqmajsfs"}
+{"time":"2026-08-14T22:24:55.662446346Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"raaqqqmajsfs"}
+{"time":"2026-08-14T22:24:56.296327526Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"raaqqqmajsfs"}
+{"time":"2026-08-14T22:24:56.297810539Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"a3607q94","id":"1(@)"}
+{"time":"2026-08-14T22:24:56.29856977Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"a3607q94","id":"1(@)"}
+{"time":"2026-08-14T22:24:56.82560196Z","level":"INFO","msg":"handleInformInit: received","streamId":"wcvy5vyw","id":"1(@)"}
+{"time":"2026-08-14T22:24:57.085772231Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"wcvy5vyw","id":"1(@)"}
+{"time":"2026-08-14T22:25:02.571913681Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"2nb8wusqu4fo"}
+{"time":"2026-08-14T22:38:28.612623637Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"2nb8wusqu4fo"}
+{"time":"2026-08-14T22:38:29.208727987Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"2nb8wusqu4fo"}
+{"time":"2026-08-14T22:38:29.210106262Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"wcvy5vyw","id":"1(@)"}
+{"time":"2026-08-14T22:38:29.210705504Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"wcvy5vyw","id":"1(@)"}
+{"time":"2026-08-14T22:38:29.839594174Z","level":"INFO","msg":"handleInformInit: received","streamId":"uuv4h936","id":"1(@)"}
+{"time":"2026-08-14T22:38:30.099444184Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"uuv4h936","id":"1(@)"}
+{"time":"2026-08-14T22:38:35.620666929Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"r0aouorh27ls"}
+{"time":"2026-08-14T22:52:17.704425638Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"r0aouorh27ls"}
+{"time":"2026-08-14T22:52:18.35448297Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"r0aouorh27ls"}
+{"time":"2026-08-14T22:52:18.355720686Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"uuv4h936","id":"1(@)"}
+{"time":"2026-08-14T22:52:18.356578583Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"uuv4h936","id":"1(@)"}
+{"time":"2026-08-14T22:52:18.899799881Z","level":"INFO","msg":"handleInformInit: received","streamId":"x87hwrtd","id":"1(@)"}
+{"time":"2026-08-14T22:52:19.159494071Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"x87hwrtd","id":"1(@)"}
+{"time":"2026-08-14T22:52:24.671083701Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"hn7lqjhi3ekb"}
+{"time":"2026-08-14T23:05:51.214192173Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"hn7lqjhi3ekb"}
+{"time":"2026-08-14T23:05:51.783107435Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"hn7lqjhi3ekb"}
+{"time":"2026-08-14T23:05:51.784410941Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"x87hwrtd","id":"1(@)"}
+{"time":"2026-08-14T23:05:51.78516667Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"x87hwrtd","id":"1(@)"}
+{"time":"2026-08-14T23:05:52.088192277Z","level":"INFO","msg":"handleInformTeardown: server teardown initiated","id":"1(@)"}
+{"time":"2026-08-14T23:05:52.088229648Z","level":"INFO","msg":"handleInformTeardown: server shutdown complete","id":"1(@)"}
+{"time":"2026-08-14T23:05:52.088239651Z","level":"INFO","msg":"connection: closing","id":"1(@)"}
+{"time":"2026-08-14T23:05:52.088243873Z","level":"INFO","msg":"server: is shutting down"}
+{"time":"2026-08-14T23:05:52.0882852Z","level":"INFO","msg":"connection: closed successfully","id":"1(@)"}
+{"time":"2026-08-14T23:05:52.088299095Z","level":"INFO","msg":"processOutgoingData: finished","id":"1(@)"}
+{"time":"2026-08-14T23:05:52.088345917Z","level":"INFO","msg":"connection: ManageConnectionData: connection closed","id":"1(@)"}
+{"time":"2026-08-14T23:05:52.088452051Z","level":"INFO","msg":"server: listener closed","addr":{"Name":"/tmp/wandb-1229423-1230766-2604858008/socket","Net":"unix"}}
+{"time":"2026-08-14T23:05:52.088520571Z","level":"INFO","msg":"server: all connections closed"}

zain/Activation/wandb/run-20260814_215902-npap5k3g/logs/debug-internal.log

@@ -0,0 +1,119 @@
+{"time":"2026-08-14T21:59:02.525025917Z","level":"INFO","msg":"wandb-core"}
+{"time":"2026-08-14T21:59:02.52517493Z","level":"INFO","msg":"stream: starting","core version":"0.28.1"}
+{"time":"2026-08-14T21:59:02.786002054Z","level":"INFO","msg":"stream: created new stream","id":"npap5k3g"}
+{"time":"2026-08-14T21:59:02.786104418Z","level":"INFO","msg":"handler: started"}
+{"time":"2026-08-14T21:59:02.786187407Z","level":"INFO","msg":"stream: started"}
+{"time":"2026-08-14T21:59:02.786198868Z","level":"INFO","msg":"writer: started","stream_id":"npap5k3g"}
+{"time":"2026-08-14T21:59:02.786231041Z","level":"INFO","msg":"sender: started"}
+{"time":"2026-08-14T21:59:03.16418738Z","level":"INFO","msg":"filestream: sending request","total_files":1,"console_offset":0,"console_lines":1}
+{"time":"2026-08-14T21:59:03.266771519Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:59:18.164192155Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":0,"events_lines":1,"console_offset":0,"console_lines":1,"uploaded_len":2}
+{"time":"2026-08-14T21:59:18.299361274Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:59:33.16463703Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":0,"history_lines":1,"events_offset":1,"events_lines":2,"console_offset":0,"console_lines":2}
+{"time":"2026-08-14T21:59:33.274514188Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T21:59:48.164822613Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":1,"history_lines":1,"events_offset":3,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-14T21:59:48.292417282Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:00:03.164276479Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":2,"history_lines":1,"events_offset":5,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-14T22:00:03.319883137Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:00:18.164816938Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":3,"history_lines":1,"events_offset":7,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-14T22:00:18.291187426Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:00:33.164651236Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":4,"history_lines":1,"events_offset":9,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-14T22:00:33.29338924Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:00:48.164357281Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":5,"history_lines":1,"events_offset":11,"events_lines":2,"console_offset":2,"console_lines":10}
+{"time":"2026-08-14T22:00:48.312694551Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:01:03.165180094Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":6,"history_lines":1,"events_offset":13,"events_lines":2,"console_offset":10,"console_lines":1}
+{"time":"2026-08-14T22:01:03.31910489Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:01:18.165111217Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":7,"history_lines":1,"events_offset":15,"events_lines":2,"console_offset":10,"console_lines":1}
+{"time":"2026-08-14T22:01:18.283466415Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:01:33.164469089Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":8,"history_lines":1,"events_offset":17,"events_lines":2,"console_offset":10,"console_lines":1}
+{"time":"2026-08-14T22:01:33.272073411Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:01:48.164574676Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":9,"history_lines":1,"events_offset":19,"events_lines":2,"console_offset":10,"console_lines":1}
+{"time":"2026-08-14T22:01:48.29257985Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:02:03.164529891Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":10,"history_lines":1,"events_offset":21,"events_lines":2,"console_offset":12,"console_lines":10}
+{"time":"2026-08-14T22:02:03.386751795Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:02:18.165330631Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":11,"history_lines":1,"events_offset":23,"events_lines":2,"console_offset":20,"console_lines":1}
+{"time":"2026-08-14T22:02:18.338577972Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:02:33.164355386Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":12,"history_lines":1,"events_offset":25,"events_lines":2,"console_offset":20,"console_lines":1}
+{"time":"2026-08-14T22:02:33.267447058Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:02:48.165019444Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":13,"history_lines":1,"events_offset":27,"events_lines":2,"console_offset":20,"console_lines":1}
+{"time":"2026-08-14T22:02:48.275154687Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:03:03.165234131Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":14,"history_lines":1,"events_offset":29,"events_lines":2,"console_offset":20,"console_lines":1}
+{"time":"2026-08-14T22:03:03.302679957Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:03:18.165058948Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":15,"history_lines":1,"events_offset":31,"events_lines":2,"console_offset":22,"console_lines":10}
+{"time":"2026-08-14T22:03:18.307459427Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:03:33.164356216Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":16,"history_lines":1,"events_offset":33,"events_lines":2,"console_offset":30,"console_lines":1}
+{"time":"2026-08-14T22:03:33.285878685Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:03:48.164932217Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":17,"history_lines":1,"events_offset":35,"events_lines":2,"console_offset":30,"console_lines":1}
+{"time":"2026-08-14T22:03:48.289134523Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:04:03.164590286Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":18,"history_lines":1,"events_offset":37,"events_lines":2,"console_offset":30,"console_lines":1}
+{"time":"2026-08-14T22:04:03.318137913Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:04:18.166174086Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":19,"history_lines":1,"events_offset":39,"events_lines":2,"console_offset":30,"console_lines":1}
+{"time":"2026-08-14T22:04:18.329722068Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:04:33.165069939Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":20,"history_lines":1,"events_offset":41,"events_lines":2,"console_offset":32,"console_lines":10}
+{"time":"2026-08-14T22:04:33.293741837Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:04:48.165112624Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":21,"history_lines":1,"events_offset":43,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-14T22:04:48.287164722Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:05:03.164528764Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":22,"history_lines":1,"events_offset":45,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-14T22:05:03.317551641Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:05:18.164433387Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":23,"history_lines":1,"events_offset":47,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-14T22:05:18.295084679Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:05:33.164912254Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":24,"history_lines":1,"events_offset":49,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-14T22:05:33.281370888Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:05:48.164748571Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":25,"history_lines":1,"events_offset":51,"events_lines":2,"console_offset":42,"console_lines":10}
+{"time":"2026-08-14T22:05:48.30493606Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:06:03.165189247Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":26,"history_lines":1,"events_offset":53,"events_lines":2,"console_offset":50,"console_lines":1}
+{"time":"2026-08-14T22:06:03.328534559Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:06:18.164737424Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":27,"history_lines":1,"events_offset":55,"events_lines":2,"console_offset":50,"console_lines":1}
+{"time":"2026-08-14T22:06:18.311417814Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:06:33.164319253Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":28,"history_lines":1,"events_offset":57,"events_lines":2,"console_offset":50,"console_lines":1}
+{"time":"2026-08-14T22:06:33.301054843Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:06:48.1644944Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":29,"history_lines":1,"events_offset":59,"events_lines":2,"console_offset":50,"console_lines":1}
+{"time":"2026-08-14T22:06:48.289979667Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:07:03.165264099Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":30,"history_lines":1,"events_offset":61,"events_lines":2,"console_offset":52,"console_lines":10}
+{"time":"2026-08-14T22:07:03.343292381Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:07:18.164973207Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":31,"history_lines":1,"events_offset":63,"events_lines":2,"console_offset":60,"console_lines":1}
+{"time":"2026-08-14T22:07:18.353573199Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:07:33.164802894Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":32,"history_lines":1,"events_offset":65,"events_lines":2,"console_offset":60,"console_lines":1}
+{"time":"2026-08-14T22:07:33.305285692Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:07:48.165189676Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":33,"history_lines":1,"events_offset":67,"events_lines":2,"console_offset":60,"console_lines":1}
+{"time":"2026-08-14T22:07:48.289133513Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:08:03.164839714Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":34,"history_lines":1,"events_offset":69,"events_lines":2,"console_offset":60,"console_lines":1}
+{"time":"2026-08-14T22:08:03.354952087Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:08:18.164344367Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":35,"history_lines":1,"events_offset":71,"events_lines":2,"console_offset":62,"console_lines":10}
+{"time":"2026-08-14T22:08:18.29928323Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:08:33.164862978Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":36,"history_lines":1,"events_offset":73,"events_lines":2,"console_offset":70,"console_lines":1}
+{"time":"2026-08-14T22:08:33.310705554Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:08:48.165076462Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":37,"history_lines":1,"events_offset":75,"events_lines":2,"console_offset":70,"console_lines":1}
+{"time":"2026-08-14T22:08:48.287531546Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:09:03.165128392Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":38,"history_lines":1,"events_offset":77,"events_lines":2,"console_offset":70,"console_lines":1}
+{"time":"2026-08-14T22:09:03.350593875Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:09:18.164398077Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":39,"history_lines":1,"events_offset":79,"events_lines":2,"console_offset":70,"console_lines":1}
+{"time":"2026-08-14T22:09:18.29071046Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:09:33.165180812Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":40,"history_lines":1,"events_offset":81,"events_lines":2,"console_offset":72,"console_lines":10}
+{"time":"2026-08-14T22:09:33.301433559Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:09:48.16448724Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":41,"history_lines":1,"events_offset":83,"events_lines":2,"console_offset":80,"console_lines":1}
+{"time":"2026-08-14T22:09:48.293952896Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:10:03.16458135Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":42,"history_lines":1,"events_offset":85,"events_lines":2,"console_offset":80,"console_lines":1}
+{"time":"2026-08-14T22:10:03.381027285Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:10:18.164382054Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":43,"history_lines":1,"events_offset":87,"events_lines":2,"console_offset":80,"console_lines":1}
+{"time":"2026-08-14T22:10:18.292431171Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:10:33.16446272Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":44,"history_lines":1,"events_offset":89,"events_lines":2,"console_offset":80,"console_lines":1}
+{"time":"2026-08-14T22:10:33.29040667Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:10:48.16469844Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":45,"history_lines":1,"events_offset":91,"events_lines":2,"console_offset":82,"console_lines":10}
+{"time":"2026-08-14T22:10:48.299861962Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:11:03.165186395Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":46,"history_lines":1,"events_offset":93,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-14T22:11:03.387650073Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:11:18.164720722Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":47,"history_lines":1,"events_offset":95,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-14T22:11:18.319368282Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:11:33.165036415Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":48,"history_lines":1,"events_offset":97,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-14T22:11:33.406468053Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:11:48.165096137Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":49,"history_lines":3,"events_offset":99,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-14T22:11:48.293634658Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:11:54.032679301Z","level":"INFO","msg":"fileTransfer: Close: file transfer manager closed"}
+{"time":"2026-08-14T22:11:54.033879099Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":52,"history_lines":1,"events_offset":101,"events_lines":1,"console_offset":92,"console_lines":18,"uploaded_len":3,"complete":true,"exit_code":0}
+{"time":"2026-08-14T22:11:54.171244354Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:11:54.172499989Z","level":"INFO","msg":"handler: operation stats","stats":{}}
+{"time":"2026-08-14T22:11:54.174925302Z","level":"INFO","msg":"stream: finishing up"}
+{"time":"2026-08-14T22:11:54.174947987Z","level":"INFO","msg":"handler: closed"}
+{"time":"2026-08-14T22:11:54.17500805Z","level":"INFO","msg":"sender: closed"}
+{"time":"2026-08-14T22:11:54.175011794Z","level":"INFO","msg":"stream: all finished"}

zain/Activation/wandb/run-20260814_215902-npap5k3g/logs/debug.log

@@ -0,0 +1,24 @@
+2026-08-14 21:59:02,523 INFO    MainThread:1229423 [wandb_init.py:setup_run_log_directory():729] Logging user logs to /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/wandb/run-20260814_215902-npap5k3g/logs/debug.log
+2026-08-14 21:59:02,523 INFO    MainThread:1229423 [wandb_init.py:setup_run_log_directory():730] Logging internal logs to /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/wandb/run-20260814_215902-npap5k3g/logs/debug-internal.log
+2026-08-14 21:59:02,523 INFO    MainThread:1229423 [wandb_init.py:init():772] calling init triggers
+2026-08-14 21:59:02,523 INFO    MainThread:1229423 [wandb_init.py:init():777] wandb.init called with sweep_config: {}
+config: {'_wandb': {}}
+2026-08-14 21:59:02,523 INFO    MainThread:1229423 [wandb_init.py:init():820] starting backend
+2026-08-14 21:59:02,523 INFO    MainThread:1229423 [wandb_init.py:init():835] sending inform_init request
+2026-08-14 21:59:02,786 INFO    MainThread:1229423 [wandb_init.py:init():840] backend started and connected
+2026-08-14 21:59:02,787 INFO    MainThread:1229423 [wandb_init.py:init():910] updated telemetry
+2026-08-14 21:59:02,794 INFO    MainThread:1229423 [wandb_init.py:init():933] communicating run to backend with 90.0 second timeout
+2026-08-14 21:59:03,083 INFO    MainThread:1229423 [wandb_init.py:init():978] starting run threads in backend
+2026-08-14 21:59:03,159 INFO    MainThread:1229423 [wandb_run.py:_console_start():2621] atexit reg
+2026-08-14 21:59:03,159 INFO    MainThread:1229423 [wandb_run.py:_redirect():2471] redirect: wrap_raw
+2026-08-14 21:59:03,159 INFO    MainThread:1229423 [wandb_run.py:_redirect():2540] Wrapping output streams.
+2026-08-14 21:59:03,159 INFO    MainThread:1229423 [wandb_run.py:_redirect():2563] Redirects installed.
+2026-08-14 21:59:03,160 INFO    MainThread:1229423 [wandb_init.py:init():1016] run started, returning control to user process
+2026-08-14 21:59:03,161 INFO    MainThread:1229423 [wandb_run.py:_config_callback():1346] config_cb None None {'transformers_version': '5.16.0.dev0', 'architectures': None, 'output_hidden_states': False, 'return_dict': True, 'dtype': None, 'chunk_size_feed_forward': 0, 'is_encoder_decoder': False, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'problem_type': None, 'vocab_size': 4096, 'hidden_size': 128, 'intermediate_size': 256, 'num_hidden_layers': 21, 'num_attention_heads': 4, 'num_key_value_heads': 4, 'hidden_act': 'silu', 'max_position_embeddings': 512, 'initializer_range': 0.02, 'rms_norm_eps': 1e-06, 'use_cache': False, 'pad_token_id': 0, 'bos_token_id': 1, 'eos_token_id': 2, 'pretraining_tp': 1, 'tie_word_embeddings': True, 'rope_parameters': {'rope_theta': 10000.0, 'rope_type': 'default'}, 'attention_bias': False, 'attention_dropout': 0.0, 'mlp_bias': False, 'head_dim': 32, '_name_or_path': '', 'tokenizer_name': 'w-ahmad/tiny-stories-tokenizer', 'mlp_type': 'mlp', 'activation': 'waleed10', 'waleed_beta': 10.0, 'model_type': 'tiny_llama', 'output_attentions': False, 'output_dir': 'out/mlp-waleed10-21L_run', 'per_device_train_batch_size': 128, 'num_train_epochs': 1, 'max_steps': 1000, 'learning_rate': 0.001, 'lr_scheduler_type': 'constant', 'lr_scheduler_kwargs': None, 'warmup_steps': 0, 'optim': 'adamw_torch', 'optim_args': None, 'weight_decay': 0.01, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'optim_target_modules': None, 'gradient_accumulation_steps': 4, 'average_tokens_across_devices': True, 'max_grad_norm': 1.0, 'label_smoothing_factor': 0.0, 'bf16': True, 'fp16': False, 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'gradient_checkpointing': False, 'gradient_checkpointing_kwargs': None, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'use_liger_kernel': False, 'liger_kernel_config': None, 'neftune_noise_alpha': None, 'torch_empty_cache_steps': None, 'auto_find_batch_size': False, 'logging_strategy': 'steps', 'logging_steps': 20, 'logging_first_step': False, 'log_on_each_node': True, 'logging_nan_inf_filter': True, 'include_num_input_tokens_seen': 'no', 'log_level': 'passive', 'log_level_replica': 'warning', 'disable_tqdm': False, 'report_to': ['wandb'], 'run_name': 'LM-mlp-waleed10-21L-4.0M-20260814-215902', 'project': 'huggingface', 'trackio_space_id': None, 'trackio_bucket_id': None, 'trackio_static_space_id': None, 'eval_strategy': 'steps', 'eval_steps': 60000, 'eval_delay': 0, 'per_device_eval_batch_size': 128, 'prediction_loss_only': False, 'eval_on_start': False, 'eval_do_concat_batches': True, 'eval_use_gather_object': False, 'eval_accumulation_steps': None, 'include_for_metrics': [], 'batch_eval_metrics': False, 'save_only_model': False, 'save_strategy': 'steps', 'save_steps': 100, 'save_on_each_node': False, 'save_total_limit': None, 'enable_jit_checkpoint': False, 'push_to_hub': False, 'hub_token': '<HUB_TOKEN>', 'hub_private_repo': None, 'hub_model_id': 'w-ahmad/A-mlp-waleed10-21L', 'hub_strategy': 'every_save', 'hub_always_push': False, 'hub_revision': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'restore_callback_states_from_checkpoint': False, 'full_determinism': False, 'seed': 42, 'data_seed': 42, 'use_cpu': False, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}, 'parallelism_config': None, 'dataloader_drop_last': False, 'dataloader_num_workers': 0, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'dataloader_prefetch_factor': None, 'dataloader_multiprocessing_context': None, 'dataloader_in_order': True, 'remove_unused_columns': False, 'label_names': None, 'train_sampling_strategy': 'random', 'length_column_name': 'length', 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'ddp_static_graph': None, 'ddp_backend': None, 'ddp_timeout': 1800, 'fsdp': None, 'fsdp_config': None, 'deepspeed': None, 'debug': [], 'skip_memory_metrics': True, 'do_train': False, 'do_eval': True, 'do_predict': False, 'resume_from_checkpoint': None, 'local_rank': -1}
+2026-08-14 21:59:03,163 INFO    MainThread:1229423 [wandb_config.py:__setitem__():155] [no run ID] config set model/num_parameters = 3970432 - <bound method Run._config_callback of <wandb.sdk.wandb_run.Run object at 0x1528e000f9d0>>
+2026-08-14 21:59:03,163 INFO    MainThread:1229423 [wandb_run.py:_config_callback():1346] config_cb model/num_parameters 3970432 None
+2026-08-14 22:11:53,642 INFO    MainThread:1229423 [wandb_run.py:_finish():2383] finishing run deepnevro-deepnevro/research1/npap5k3g
+2026-08-14 22:11:53,643 INFO    MainThread:1229423 [wandb_run.py:_atexit_cleanup():2588] got exitcode: 0
+2026-08-14 22:11:53,643 INFO    MainThread:1229423 [wandb_run.py:_restore():2570] restore
+2026-08-14 22:11:53,643 INFO    MainThread:1229423 [wandb_run.py:_restore():2576] restore done
+2026-08-14 22:11:54,174 INFO    MainThread:1229423 [wandb_run.py:_footer_sync_info():3993] logging synced files

zain/Activation/wandb/run-20260814_215902-npap5k3g/run-npap5k3g.wandb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a6ad8a8f597b0c7341cdd99b8a47d1d3d1be062d82972d78c7311251214c50c5
+size 563606

zain/Activation/wandb/run-20260814_221154-a3607q94/files/config.yaml

@@ -0,0 +1,439 @@
+_name_or_path:
+    value: ""
+_wandb:
+    value:
+        cli_version: 0.28.1
+        e:
+            uq8vhfi992gapn2qqlagfoc9r0woeq20:
+                args:
+                    - --config
+                    - /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/configs/baseline.yaml
+                    - --variants
+                    - glu-waleed10
+                    - glu-silu-waleed10
+                    - mlp-waleed10
+                    - mlp-silu-waleed10
+                    - glu-waleed
+                    - glu-situglu_low
+                    - glu-waleedglu_low
+                codePath: sweep.py
+                codePathLocal: sweep.py
+                cpu_count: 112
+                cpu_count_logical: 224
+                cudaVersion: "12.4"
+                disk:
+                    /:
+                        total: "1560765693952"
+                        used: "709095636992"
+                email: deepnevro@gmail.com
+                executable: /mnt/data/zainulabideen/zain-exp/notebooks/my_env/bin/python
+                git:
+                    commit: 3fb47c6943d6927eafc39bd399a3d3a520bae74a
+                    remote: https://github.com/w-ahmad1a10/Activation.git
+                gpu: NVIDIA H100 80GB HBM3
+                gpu_count: 8
+                gpu_nvidia:
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-39c684a5-fde6-83d7-1663-0859795881ae
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-68012e5a-38b6-b643-0ca6-62fb66720bf3
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-132944c4-b689-2b5f-89a4-d730401677ab
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-2df386cc-6d26-d0e2-7a2d-a057b0d95864
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-bfa16575-1d94-1aa2-4537-2c93433f42ef
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-bc6c3e3c-9b90-09ca-c034-774961847c54
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-00a441e1-7c95-e7d6-4c35-43d6b291aea9
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-1c4d29a2-4647-6fce-d8fc-0c5ecfbbd6ea
+                host: deeplens-k3s-node1
+                memory:
+                    total: "2164089937920"
+                os: Linux-5.15.0-126-generic-x86_64-with-glibc2.35
+                program: /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/sweep.py
+                python: CPython 3.11.15
+                root: /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation
+                startedAt: "2026-08-14T22:11:54.692447Z"
+                writerId: uq8vhfi992gapn2qqlagfoc9r0woeq20
+        m:
+            - "1": train/global_step
+              "6":
+                - 3
+              "7": []
+            - "2": '*'
+              "5": 1
+              "6":
+                - 1
+              "7": []
+        python_version: 3.11.15
+        t:
+            "1":
+                - 1
+                - 5
+                - 11
+                - 41
+                - 49
+                - 51
+                - 53
+                - 71
+            "2":
+                - 1
+                - 5
+                - 11
+                - 41
+                - 49
+                - 51
+                - 53
+                - 71
+            "3":
+                - 2
+                - 7
+                - 13
+                - 19
+                - 62
+                - 66
+            "4": 3.11.15
+            "5": 0.28.1
+            "6": 5.16.0.dev0
+            "9":
+                "1": transformers_trainer
+            "12": 0.28.1
+            "13": linux-x86_64
+accelerator_config:
+    value:
+        dispatch_batches: null
+        even_batches: true
+        gradient_accumulation_kwargs: null
+        non_blocking: false
+        split_batches: false
+        use_seedable_sampler: true
+activation:
+    value: silu-waleed10
+adam_beta1:
+    value: 0.9
+adam_beta2:
+    value: 0.999
+adam_epsilon:
+    value: 1e-08
+architectures:
+    value: null
+attention_bias:
+    value: false
+attention_dropout:
+    value: 0
+auto_find_batch_size:
+    value: false
+average_tokens_across_devices:
+    value: true
+batch_eval_metrics:
+    value: false
+bf16:
+    value: true
+bf16_full_eval:
+    value: false
+bos_token_id:
+    value: 1
+chunk_size_feed_forward:
+    value: 0
+data_seed:
+    value: 42
+dataloader_drop_last:
+    value: false
+dataloader_in_order:
+    value: true
+dataloader_multiprocessing_context:
+    value: null
+dataloader_num_workers:
+    value: 0
+dataloader_persistent_workers:
+    value: false
+dataloader_pin_memory:
+    value: true
+dataloader_prefetch_factor:
+    value: null
+ddp_backend:
+    value: null
+ddp_broadcast_buffers:
+    value: null
+ddp_bucket_cap_mb:
+    value: null
+ddp_find_unused_parameters:
+    value: null
+ddp_static_graph:
+    value: null
+ddp_timeout:
+    value: 1800
+debug:
+    value: []
+deepspeed:
+    value: null
+disable_tqdm:
+    value: false
+do_eval:
+    value: true
+do_predict:
+    value: false
+do_train:
+    value: false
+dtype:
+    value: null
+enable_jit_checkpoint:
+    value: false
+eos_token_id:
+    value: 2
+eval_accumulation_steps:
+    value: null
+eval_delay:
+    value: 0
+eval_do_concat_batches:
+    value: true
+eval_on_start:
+    value: false
+eval_steps:
+    value: 60000
+eval_strategy:
+    value: steps
+eval_use_gather_object:
+    value: false
+fp16:
+    value: false
+fp16_full_eval:
+    value: false
+fsdp:
+    value: null
+fsdp_config:
+    value: null
+full_determinism:
+    value: false
+gradient_accumulation_steps:
+    value: 4
+gradient_checkpointing:
+    value: false
+gradient_checkpointing_kwargs:
+    value: null
+greater_is_better:
+    value: null
+head_dim:
+    value: 32
+hidden_act:
+    value: silu
+hidden_size:
+    value: 128
+hub_always_push:
+    value: false
+hub_model_id:
+    value: w-ahmad/A-mlp-silu-waleed10-21L
+hub_private_repo:
+    value: null
+hub_revision:
+    value: null
+hub_strategy:
+    value: every_save
+hub_token:
+    value: <HUB_TOKEN>
+id2label:
+    value:
+        "0": LABEL_0
+        "1": LABEL_1
+ignore_data_skip:
+    value: false
+include_for_metrics:
+    value: []
+include_num_input_tokens_seen:
+    value: "no"
+initializer_range:
+    value: 0.02
+intermediate_size:
+    value: 256
+is_encoder_decoder:
+    value: false
+label_names:
+    value: null
+label_smoothing_factor:
+    value: 0
+label2id:
+    value:
+        LABEL_0: 0
+        LABEL_1: 1
+learning_rate:
+    value: 0.001
+length_column_name:
+    value: length
+liger_kernel_config:
+    value: null
+load_best_model_at_end:
+    value: false
+local_rank:
+    value: -1
+log_level:
+    value: passive
+log_level_replica:
+    value: warning
+log_on_each_node:
+    value: true
+logging_first_step:
+    value: false
+logging_nan_inf_filter:
+    value: true
+logging_steps:
+    value: 20
+logging_strategy:
+    value: steps
+lr_scheduler_kwargs:
+    value: null
+lr_scheduler_type:
+    value: constant
+max_grad_norm:
+    value: 1
+max_position_embeddings:
+    value: 512
+max_steps:
+    value: 1000
+metric_for_best_model:
+    value: null
+mlp_bias:
+    value: false
+mlp_type:
+    value: mlp
+model/num_parameters:
+    value: 3970432
+model_type:
+    value: tiny_llama
+neftune_noise_alpha:
+    value: null
+num_attention_heads:
+    value: 4
+num_hidden_layers:
+    value: 21
+num_key_value_heads:
+    value: 4
+num_train_epochs:
+    value: 1
+optim:
+    value: adamw_torch
+optim_args:
+    value: null
+optim_target_modules:
+    value: null
+output_attentions:
+    value: false
+output_dir:
+    value: out/mlp-silu-waleed10-21L_run
+output_hidden_states:
+    value: false
+pad_token_id:
+    value: 0
+parallelism_config:
+    value: null
+per_device_eval_batch_size:
+    value: 128
+per_device_train_batch_size:
+    value: 128
+prediction_loss_only:
+    value: false
+pretraining_tp:
+    value: 1
+problem_type:
+    value: null
+project:
+    value: huggingface
+push_to_hub:
+    value: false
+remove_unused_columns:
+    value: false
+report_to:
+    value:
+        - wandb
+restore_callback_states_from_checkpoint:
+    value: false
+resume_from_checkpoint:
+    value: null
+return_dict:
+    value: true
+rms_norm_eps:
+    value: 1e-06
+rope_parameters:
+    value:
+        rope_theta: 10000
+        rope_type: default
+run_name:
+    value: LM-mlp-silu-waleed10-21L-4.0M-20260814-221154
+save_on_each_node:
+    value: false
+save_only_model:
+    value: false
+save_steps:
+    value: 100
+save_strategy:
+    value: steps
+save_total_limit:
+    value: null
+seed:
+    value: 42
+skip_memory_metrics:
+    value: true
+tf32:
+    value: null
+tie_word_embeddings:
+    value: true
+tokenizer_name:
+    value: w-ahmad/tiny-stories-tokenizer
+torch_compile:
+    value: false
+torch_compile_backend:
+    value: null
+torch_compile_mode:
+    value: null
+torch_empty_cache_steps:
+    value: null
+trackio_bucket_id:
+    value: null
+trackio_space_id:
+    value: null
+trackio_static_space_id:
+    value: null
+train_sampling_strategy:
+    value: random
+transformers_version:
+    value: 5.16.0.dev0
+use_cache:
+    value: false
+use_cpu:
+    value: false
+use_liger_kernel:
+    value: false
+vocab_size:
+    value: 4096
+waleed_beta:
+    value: 10
+warmup_steps:
+    value: 0
+weight_decay:
+    value: 0.01

zain/Activation/wandb/run-20260814_221154-a3607q94/files/output.log

@@ -0,0 +1,110 @@
+ 10%|█         | 100/1000 [01:16<11:17,  1.33it/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': '29.1', 'grad_norm': '4.094', 'learning_rate': '0.001', 'epoch': '0.01078', 'train/total_time_seconds': '7.585', 'train/time_per_step_avg': '0.3793', 'train/epoch_time_elapsed': '15.48', 'train/estimated_remaining_minutes': '6.194'}
+{'loss': '24.32', 'grad_norm': '2.219', 'learning_rate': '0.001', 'epoch': '0.02157', 'train/total_time_seconds': '15.13', 'train/time_per_step_avg': '0.3783', 'train/epoch_time_elapsed': '30.63', 'train/estimated_remaining_minutes': '6.053'}
+{'loss': '22.97', 'grad_norm': '1.391', 'learning_rate': '0.001', 'epoch': '0.03235', 'train/total_time_seconds': '22.67', 'train/time_per_step_avg': '0.3779', 'train/epoch_time_elapsed': '45.83', 'train/estimated_remaining_minutes': '5.92'}
+{'loss': '21.97', 'grad_norm': '1.625', 'learning_rate': '0.001', 'epoch': '0.04314', 'train/total_time_seconds': '30.22', 'train/time_per_step_avg': '0.3777', 'train/epoch_time_elapsed': '60.9', 'train/estimated_remaining_minutes': '5.792'}
+{'loss': '20.95', 'grad_norm': '3.625', 'learning_rate': '0.001', 'epoch': '0.05392', 'train/total_time_seconds': '37.75', 'train/time_per_step_avg': '0.3775', 'train/epoch_time_elapsed': '76.14', 'train/estimated_remaining_minutes': '5.662'}
+  - 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, 47.25it/s]
+ 20%|██        | 200/1000 [02:31<09:58,  1.34it/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': '20.18', 'grad_norm': '3.375', 'learning_rate': '0.001', 'epoch': '0.06471', 'train/total_time_seconds': '45.25', 'train/time_per_step_avg': '0.3767', 'train/epoch_time_elapsed': '91.24', 'train/estimated_remaining_minutes': '5.531'}
+{'loss': '19.39', 'grad_norm': '3.797', 'learning_rate': '0.001', 'epoch': '0.07549', 'train/total_time_seconds': '52.77', 'train/time_per_step_avg': '0.3763', 'train/epoch_time_elapsed': '106.3', 'train/estimated_remaining_minutes': '5.402'}
+{'loss': '18.66', 'grad_norm': '3.219', 'learning_rate': '0.001', 'epoch': '0.08628', 'train/total_time_seconds': '60.34', 'train/time_per_step_avg': '0.3767', 'train/epoch_time_elapsed': '121.4', 'train/estimated_remaining_minutes': '5.28'}
+{'loss': '17.96', 'grad_norm': '2.609', 'learning_rate': '0.001', 'epoch': '0.09706', 'train/total_time_seconds': '67.9', 'train/time_per_step_avg': '0.3768', 'train/epoch_time_elapsed': '136.7', 'train/estimated_remaining_minutes': '5.155'}
+{'loss': '17.32', 'grad_norm': '2.359', 'learning_rate': '0.001', 'epoch': '0.1078', 'train/total_time_seconds': '75.42', 'train/time_per_step_avg': '0.3767', 'train/epoch_time_elapsed': '151.7', 'train/estimated_remaining_minutes': '5.028'}
+  - 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, 85.88it/s]
+ 30%|███       | 300/1000 [03:47<09:14,  1.26it/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': '16.67', 'grad_norm': '3.484', 'learning_rate': '0.001', 'epoch': '0.1186', 'train/total_time_seconds': '82.96', 'train/time_per_step_avg': '0.3771', 'train/epoch_time_elapsed': '166.9', 'train/estimated_remaining_minutes': '4.902'}
+{'loss': '16.15', 'grad_norm': '4.062', 'learning_rate': '0.001', 'epoch': '0.1294', 'train/total_time_seconds': '90.52', 'train/time_per_step_avg': '0.3775', 'train/epoch_time_elapsed': '182', 'train/estimated_remaining_minutes': '4.777'}
+{'loss': '15.67', 'grad_norm': '2.344', 'learning_rate': '0.001', 'epoch': '0.1402', 'train/total_time_seconds': '98.09', 'train/time_per_step_avg': '0.3775', 'train/epoch_time_elapsed': '197.2', 'train/estimated_remaining_minutes': '4.653'}
+{'loss': '15.26', 'grad_norm': '2.984', 'learning_rate': '0.001', 'epoch': '0.151', 'train/total_time_seconds': '105.6', 'train/time_per_step_avg': '0.3774', 'train/epoch_time_elapsed': '212.4', 'train/estimated_remaining_minutes': '4.527'}
+{'loss': '14.91', 'grad_norm': '1.812', 'learning_rate': '0.001', 'epoch': '0.1618', 'train/total_time_seconds': '113.2', 'train/time_per_step_avg': '0.3778', 'train/epoch_time_elapsed': '227.8', 'train/estimated_remaining_minutes': '4.402'}
+  - 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, 67.77it/s]
+ 40%|████      | 400/1000 [05:04<07:29,  1.33it/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': '14.61', 'grad_norm': '2.391', 'learning_rate': '0.001', 'epoch': '0.1726', 'train/total_time_seconds': '120.8', 'train/time_per_step_avg': '0.3781', 'train/epoch_time_elapsed': '243.2', 'train/estimated_remaining_minutes': '4.277'}
+{'loss': '14.27', 'grad_norm': '1.648', 'learning_rate': '0.001', 'epoch': '0.1833', 'train/total_time_seconds': '128.3', 'train/time_per_step_avg': '0.3783', 'train/epoch_time_elapsed': '258.5', 'train/estimated_remaining_minutes': '4.152'}
+{'loss': '13.94', 'grad_norm': '2.031', 'learning_rate': '0.001', 'epoch': '0.1941', 'train/total_time_seconds': '135.9', 'train/time_per_step_avg': '0.3781', 'train/epoch_time_elapsed': '273.8', 'train/estimated_remaining_minutes': '4.027'}
+{'loss': '13.63', 'grad_norm': '2.75', 'learning_rate': '0.001', 'epoch': '0.2049', 'train/total_time_seconds': '143.5', 'train/time_per_step_avg': '0.3784', 'train/epoch_time_elapsed': '289.1', 'train/estimated_remaining_minutes': '3.902'}
+{'loss': '13.42', 'grad_norm': '2.688', 'learning_rate': '0.001', 'epoch': '0.2157', 'train/total_time_seconds': '151', 'train/time_per_step_avg': '0.3778', 'train/epoch_time_elapsed': '304', 'train/estimated_remaining_minutes': '3.774'}
+  - 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, 89.52it/s]
+ 50%|█████     | 500/1000 [06:20<06:18,  1.32it/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': '13.16', 'grad_norm': '2.641', 'learning_rate': '0.001', 'epoch': '0.2265', 'train/total_time_seconds': '158.5', 'train/time_per_step_avg': '0.3777', 'train/epoch_time_elapsed': '319.4', 'train/estimated_remaining_minutes': '3.649'}
+{'loss': '12.97', 'grad_norm': '2.031', 'learning_rate': '0.001', 'epoch': '0.2373', 'train/total_time_seconds': '166.1', 'train/time_per_step_avg': '0.3775', 'train/epoch_time_elapsed': '334.5', 'train/estimated_remaining_minutes': '3.523'}
+{'loss': '12.77', 'grad_norm': '2.375', 'learning_rate': '0.001', 'epoch': '0.248', 'train/total_time_seconds': '173.7', 'train/time_per_step_avg': '0.3776', 'train/epoch_time_elapsed': '349.6', 'train/estimated_remaining_minutes': '3.398'}
+{'loss': '12.56', 'grad_norm': '1.961', 'learning_rate': '0.001', 'epoch': '0.2588', 'train/total_time_seconds': '181.2', 'train/time_per_step_avg': '0.3775', 'train/epoch_time_elapsed': '364.9', 'train/estimated_remaining_minutes': '3.272'}
+{'loss': '12.36', 'grad_norm': '1.664', 'learning_rate': '0.001', 'epoch': '0.2696', 'train/total_time_seconds': '188.8', 'train/time_per_step_avg': '0.3782', 'train/epoch_time_elapsed': '380.1', 'train/estimated_remaining_minutes': '3.147'}
+  - 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, 79.83it/s]
+ 60%|██████    | 600/1000 [07:36<05:01,  1.33it/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': '12.23', 'grad_norm': '2.406', 'learning_rate': '0.001', 'epoch': '0.2804', 'train/total_time_seconds': '196.4', 'train/time_per_step_avg': '0.3783', 'train/epoch_time_elapsed': '395.5', 'train/estimated_remaining_minutes': '3.021'}
+{'loss': '12.07', 'grad_norm': '2.188', 'learning_rate': '0.001', 'epoch': '0.2912', 'train/total_time_seconds': '203.9', 'train/time_per_step_avg': '0.3782', 'train/epoch_time_elapsed': '410.7', 'train/estimated_remaining_minutes': '2.895'}
+{'loss': '11.92', 'grad_norm': '2.25', 'learning_rate': '0.001', 'epoch': '0.302', 'train/total_time_seconds': '211.4', 'train/time_per_step_avg': '0.3777', 'train/epoch_time_elapsed': '425.8', 'train/estimated_remaining_minutes': '2.769'}
+{'loss': '11.78', 'grad_norm': '2.062', 'learning_rate': '0.001', 'epoch': '0.3128', 'train/total_time_seconds': '218.9', 'train/time_per_step_avg': '0.3772', 'train/epoch_time_elapsed': '440.9', 'train/estimated_remaining_minutes': '2.642'}
+{'loss': '11.67', 'grad_norm': '2.562', 'learning_rate': '0.001', 'epoch': '0.3235', 'train/total_time_seconds': '226.5', 'train/time_per_step_avg': '0.3767', 'train/epoch_time_elapsed': '456.3', 'train/estimated_remaining_minutes': '2.516'}
+  - 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, 78.46it/s]
+ 70%|███████   | 700/1000 [08:52<03:45,  1.33it/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': '11.55', 'grad_norm': '1.992', 'learning_rate': '0.001', 'epoch': '0.3343', 'train/total_time_seconds': '234', 'train/time_per_step_avg': '0.3766', 'train/epoch_time_elapsed': '471.7', 'train/estimated_remaining_minutes': '2.391'}
+{'loss': '11.43', 'grad_norm': '2', 'learning_rate': '0.001', 'epoch': '0.3451', 'train/total_time_seconds': '241.6', 'train/time_per_step_avg': '0.3767', 'train/epoch_time_elapsed': '486.9', 'train/estimated_remaining_minutes': '2.265'}
+{'loss': '11.32', 'grad_norm': '2.203', 'learning_rate': '0.001', 'epoch': '0.3559', 'train/total_time_seconds': '249.1', 'train/time_per_step_avg': '0.377', 'train/epoch_time_elapsed': '502.2', 'train/estimated_remaining_minutes': '2.139'}
+{'loss': '11.2', 'grad_norm': '2.156', 'learning_rate': '0.001', 'epoch': '0.3667', 'train/total_time_seconds': '256.7', 'train/time_per_step_avg': '0.3775', 'train/epoch_time_elapsed': '517.4', 'train/estimated_remaining_minutes': '2.013'}
+{'loss': '11.1', 'grad_norm': '2.422', 'learning_rate': '0.001', 'epoch': '0.3775', 'train/total_time_seconds': '264.3', 'train/time_per_step_avg': '0.3779', 'train/epoch_time_elapsed': '532.6', 'train/estimated_remaining_minutes': '1.888'}
+  - 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, 47.34it/s]
+ 80%|████████  | 800/1000 [10:08<02:32,  1.31it/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': '10.98', 'grad_norm': '2.031', 'learning_rate': '0.001', 'epoch': '0.3882', 'train/total_time_seconds': '271.8', 'train/time_per_step_avg': '0.378', 'train/epoch_time_elapsed': '547.8', 'train/estimated_remaining_minutes': '1.762'}
+{'loss': '10.93', 'grad_norm': '2.781', 'learning_rate': '0.001', 'epoch': '0.399', 'train/total_time_seconds': '279.4', 'train/time_per_step_avg': '0.378', 'train/epoch_time_elapsed': '563.1', 'train/estimated_remaining_minutes': '1.636'}
+{'loss': '10.84', 'grad_norm': '2.266', 'learning_rate': '0.001', 'epoch': '0.4098', 'train/total_time_seconds': '286.9', 'train/time_per_step_avg': '0.3777', 'train/epoch_time_elapsed': '578.3', 'train/estimated_remaining_minutes': '1.51'}
+{'loss': '10.76', 'grad_norm': '2.531', 'learning_rate': '0.001', 'epoch': '0.4206', 'train/total_time_seconds': '294.5', 'train/time_per_step_avg': '0.3779', 'train/epoch_time_elapsed': '593.6', 'train/estimated_remaining_minutes': '1.384'}
+{'loss': '10.63', 'grad_norm': '2.375', 'learning_rate': '0.001', 'epoch': '0.4314', 'train/total_time_seconds': '302', 'train/time_per_step_avg': '0.3777', 'train/epoch_time_elapsed': '609', 'train/estimated_remaining_minutes': '1.258'}
+  - 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, 64.95it/s]
+ 90%|█████████ | 900/1000 [11:25<01:18,  1.27it/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': '10.59', 'grad_norm': '2.469', 'learning_rate': '0.001', 'epoch': '0.4422', 'train/total_time_seconds': '309.6', 'train/time_per_step_avg': '0.3775', 'train/epoch_time_elapsed': '624.2', 'train/estimated_remaining_minutes': '1.133'}
+{'loss': '10.52', 'grad_norm': '2.531', 'learning_rate': '0.001', 'epoch': '0.453', 'train/total_time_seconds': '317.1', 'train/time_per_step_avg': '0.3776', 'train/epoch_time_elapsed': '639.5', 'train/estimated_remaining_minutes': '1.007'}
+{'loss': '10.43', 'grad_norm': '2.484', 'learning_rate': '0.001', 'epoch': '0.4637', 'train/total_time_seconds': '324.7', 'train/time_per_step_avg': '0.3777', 'train/epoch_time_elapsed': '654.6', 'train/estimated_remaining_minutes': '0.8809'}
+{'loss': '10.36', 'grad_norm': '2.234', 'learning_rate': '0.001', 'epoch': '0.4745', 'train/total_time_seconds': '332.2', 'train/time_per_step_avg': '0.3775', 'train/epoch_time_elapsed': '670', 'train/estimated_remaining_minutes': '0.7551'}
+{'loss': '10.3', 'grad_norm': '2.641', 'learning_rate': '0.001', 'epoch': '0.4853', 'train/total_time_seconds': '339.8', 'train/time_per_step_avg': '0.3778', 'train/epoch_time_elapsed': '685.3', 'train/estimated_remaining_minutes': '0.6293'}
+  - 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, 82.06it/s]
+100%|██████████| 1000/1000 [12:50<00:00,  1.28i[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': '10.23', 'grad_norm': '2.703', 'learning_rate': '0.001', 'epoch': '0.4961', 'train/total_time_seconds': '347.4', 'train/time_per_step_avg': '0.3777', 'train/epoch_time_elapsed': '700.6', 'train/estimated_remaining_minutes': '0.5034'}
+{'loss': '10.2', 'grad_norm': '2.984', 'learning_rate': '0.001', 'epoch': '0.5069', 'train/total_time_seconds': '354.9', 'train/time_per_step_avg': '0.3773', 'train/epoch_time_elapsed': '715.6', 'train/estimated_remaining_minutes': '0.3775'}
+{'loss': '10.15', 'grad_norm': '2.234', 'learning_rate': '0.001', 'epoch': '0.5177', 'train/total_time_seconds': '362.4', 'train/time_per_step_avg': '0.377', 'train/epoch_time_elapsed': '730.9', 'train/estimated_remaining_minutes': '0.2516'}
+{'loss': '10.11', 'grad_norm': '2.469', 'learning_rate': '0.001', 'epoch': '0.5284', 'train/total_time_seconds': '369.9', 'train/time_per_step_avg': '0.3765', 'train/epoch_time_elapsed': '746', 'train/estimated_remaining_minutes': '0.1258'}
+{'loss': '10.05', 'grad_norm': '2.609', 'learning_rate': '0.001', 'epoch': '0.5392', 'train/total_time_seconds': '377.4', 'train/time_per_step_avg': '0.376', 'train/epoch_time_elapsed': '761.2', 'train/estimated_remaining_minutes': '0'}
+  - 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': '2.505', 'eval_runtime': '9.414', 'eval_samples_per_second': '1012', 'eval_steps_per_second': '7.967', 'epoch': '0.5392', 'train/total_time_seconds': '377.4', 'train/time_per_step_avg': '0.376', 'train/epoch_time_elapsed': '770.6', 'train/estimated_remaining_minutes': '0'}
+  - 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, 78.21it/s]
+100%|██████████| 1000/1000 [12:50<00:00,  1.30it/s], ?it/s]
+{'train_runtime': '771.5', 'train_samples_per_second': '663.7', 'train_steps_per_second': '1.296', 'train_loss': '13.97', 'epoch': '0.5392', 'train/total_time_seconds': '377.4', 'train/time_per_step_avg': '0.376', 'train/epoch_time_elapsed': '770.7', 'train/estimated_remaining_minutes': '0'}
+100%|██████████| 75/75 [00:09<00:00,  8.05it/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.
+  - 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, 81.46it/s]
+>>> FINISHED mlp-silu-waleed10-21L successfully

zain/Activation/wandb/run-20260814_221154-a3607q94/files/requirements.txt

@@ -0,0 +1,149 @@
+asttokens==3.0.1
+comm==0.2.3
+debugpy==1.8.21
+decorator==5.3.1
+executing==2.2.1
+nest-asyncio==1.6.0
+parso==0.8.7
+platformdirs==4.11.0
+psutil==7.2.2
+ptyprocess==0.7.0
+pure_eval==0.2.3
+Pygments==2.20.0
+pyzmq==27.1.0
+setuptools==83.0.0
+six==1.17.0
+tornado==6.5.7
+traitlets==5.15.0
+fsspec==2026.4.0
+wcwidth==0.8.2
+ipython_pygments_lexers==1.1.1
+jedi==0.20.0
+jupyter_core==5.9.1
+matplotlib-inline==0.2.2
+pexpect==4.9.0
+prompt_toolkit==3.0.53
+python-dateutil==2.9.0.post0
+stack_data==0.6.3
+wheel==0.47.0
+jupyter_client==8.9.1
+pip==26.1.2
+ipython==9.15.0
+ipykernel==7.2.0
+threadpoolctl==3.6.0
+pyparsing==3.3.2
+typing_extensions==4.15.0
+Jinja2==3.1.6
+narwhals==2.24.0
+kiwisolver==1.5.0
+joblib==1.5.3
+fonttools==4.63.0
+cycler==0.12.1
+scipy==1.17.1
+pandas==3.0.5
+contourpy==1.3.3
+scikit-learn==1.9.0
+matplotlib==3.11.1
+urllib3==2.7.0
+tqdm==4.70.0
+idna==3.18
+charset-normalizer==3.4.9
+certifi==2026.7.22
+requests==2.34.2
+seaborn==0.13.2
+uv==0.12.0
+shellingham==1.5.4
+mpmath==1.3.0
+attrs==26.1.0
+hf-xet==1.5.2
+nvidia-nccl-cu12==2.21.5
+MarkupSafe==3.0.3
+regex==2026.7.19
+importlib_metadata==9.0.0
+httpcore==1.0.9
+annotated-doc==0.0.5
+multidict==6.7.1
+aiohttp==3.14.3
+aiosignal==1.4.0
+xxhash==3.8.1
+aiohappyeyeballs==2.7.1
+mdurl==0.1.2
+cuda-toolkit==13.0.3.0
+networkx==3.6.1
+PyYAML==6.0.3
+nvidia-cufile==1.15.1.6
+typer==0.27.0
+torchaudio==2.6.0+cu124
+rich==15.0.0
+nvidia-cufft-cu12==11.2.1.3
+h11==0.16.0
+dill==0.4.1
+cuda-pathfinder==1.6.0
+filelock==3.29.0
+nvidia-nvtx-cu12==12.4.127
+httpx==0.28.1
+anyio==4.14.2
+numpy==2.4.4
+yarl==1.24.5
+click==8.4.2
+triton==3.2.0
+frozenlist==1.8.0
+zipp==4.1.0
+propcache==0.5.2
+tokenizers==0.22.2
+markdown-it-py==4.2.0
+nvidia-cuda-runtime==13.0.96
+cuda-bindings==13.3.1
+nvidia-cuda-cupti==13.0.85
+torch==2.6.0+cu124
+multiprocess==0.70.19
+pillow==12.2.0
+transformers==5.16.0.dev0
+wandb==0.28.1
+nvidia-curand==10.4.0.35
+sympy==1.13.1
+nvidia-cusparse==12.6.3.3
+nvidia-cuda-nvrtc==13.0.88
+typing-inspection==0.4.2
+nvidia-cusolver==12.0.4.66
+nvidia-cufft==12.0.0.61
+nvidia-cudnn-cu13==9.20.0.48
+nvidia-cublas==13.1.1.3
+pyarrow==25.0.0
+evaluate==0.4.6
+diffusers==0.39.0
+pydantic==2.13.4
+annotated-types==0.8.0
+protobuf==7.35.1
+sentry-sdk==2.66.1
+einops==0.8.2
+packaging==26.2
+nvidia-nvjitlink-cu12==12.4.127
+nvidia-curand-cu12==10.3.5.147
+nvidia-cusparselt-cu12==0.6.2
+nvidia-cusparse-cu12==12.3.1.170
+nvidia-cuda-runtime-cu12==12.4.127
+torchvision==0.21.0+cu124
+nvidia-cuda-nvrtc-cu12==12.4.127
+nvidia-cuda-cupti-cu12==12.4.127
+nvidia-cusolver-cu12==11.6.1.9
+nvidia-cublas-cu12==12.4.5.8
+nvidia-cudnn-cu12==9.1.0.70
+huggingface_hub==1.26.0
+datasets==5.0.1
+safetensors==0.8.0
+accelerate==1.14.0
+pydantic_core==2.46.4
+ninja==1.13.0
+autocommand==2.2.2
+backports.tarfile==1.2.0
+importlib_metadata==8.7.1
+jaraco.text==4.0.0
+jaraco.context==6.1.0
+jaraco.functools==4.4.0
+more-itertools==10.8.0
+packaging==26.0
+platformdirs==4.4.0
+tomli==2.4.0
+wheel==0.46.3
+zipp==3.23.0

zain/Activation/wandb/run-20260814_221154-a3607q94/files/wandb-metadata.json

@@ -0,0 +1,101 @@
+{
+  "os":  "Linux-5.15.0-126-generic-x86_64-with-glibc2.35",
+  "python":  "CPython 3.11.15",
+  "startedAt":  "2026-08-14T22:11:54.692447Z",
+  "args":  [
+    "--config",
+    "/mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/configs/baseline.yaml",
+    "--variants",
+    "glu-waleed10",
+    "glu-silu-waleed10",
+    "mlp-waleed10",
+    "mlp-silu-waleed10",
+    "glu-waleed",
+    "glu-situglu_low",
+    "glu-waleedglu_low"
+  ],
+  "program":  "/mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/sweep.py",
+  "codePath":  "sweep.py",
+  "codePathLocal":  "sweep.py",
+  "git":  {
+    "remote":  "https://github.com/w-ahmad1a10/Activation.git",
+    "commit":  "3fb47c6943d6927eafc39bd399a3d3a520bae74a"
+  },
+  "email":  "deepnevro@gmail.com",
+  "root":  "/mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation",
+  "host":  "deeplens-k3s-node1",
+  "executable":  "/mnt/data/zainulabideen/zain-exp/notebooks/my_env/bin/python",
+  "cpu_count":  112,
+  "cpu_count_logical":  224,
+  "gpu":  "NVIDIA H100 80GB HBM3",
+  "gpu_count":  8,
+  "disk":  {
+    "/":  {
+      "total":  "1560765693952",
+      "used":  "709095636992"
+    }
+  },
+  "memory":  {
+    "total":  "2164089937920"
+  },
+  "gpu_nvidia":  [
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-39c684a5-fde6-83d7-1663-0859795881ae"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-68012e5a-38b6-b643-0ca6-62fb66720bf3"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-132944c4-b689-2b5f-89a4-d730401677ab"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-2df386cc-6d26-d0e2-7a2d-a057b0d95864"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-bfa16575-1d94-1aa2-4537-2c93433f42ef"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-bc6c3e3c-9b90-09ca-c034-774961847c54"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-00a441e1-7c95-e7d6-4c35-43d6b291aea9"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-1c4d29a2-4647-6fce-d8fc-0c5ecfbbd6ea"
+    }
+  ],
+  "cudaVersion":  "12.4",
+  "writerId":  "uq8vhfi992gapn2qqlagfoc9r0woeq20"
+}

zain/Activation/wandb/run-20260814_221154-a3607q94/files/wandb-summary.json

@@ -0,0 +1 @@
+{"total_flos":5.420315836416e+15,"train_samples_per_second":663.654,"eval/steps_per_second":7.88,"train_runtime":771.486,"train_loss":13.971002395629883,"eval/runtime":9.5175,"_wandb":{"runtime":780},"train/global_step":1000,"train/loss":10.053606414794922,"_timestamp":1.7867462956361027e+09,"eval/loss":2.5052456855773926,"train/learning_rate":0.001,"_step":52,"train/grad_norm":2.609375,"_runtime":780,"eval/samples_per_second":1001.003,"train_steps_per_second":1.296,"train/epoch":0.5392289026691831}

zain/Activation/wandb/run-20260814_221154-a3607q94/logs/debug-core.log

@@ -0,0 +1,62 @@
+{"time":"2026-08-14T21:31:53.539152008Z","level":"INFO","msg":"main: starting server","port-filename":"/tmp/tmp650y3sc2/port-1229423.txt","pid":1229423,"detached":false,"idle-timeout":600000000000,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false,"enable-dcgm-profiling":false}
+{"time":"2026-08-14T21:31:53.540676876Z","level":"INFO","msg":"server: will exit if parent process dies","ppid":1229423}
+{"time":"2026-08-14T21:31:53.540659035Z","level":"INFO","msg":"server: accepting connections","addr":{"Name":"/tmp/wandb-1229423-1230766-2604858008/socket","Net":"unix"}}
+{"time":"2026-08-14T21:31:53.717724383Z","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"1(@)"}
+{"time":"2026-08-14T21:31:53.802092564Z","level":"INFO","msg":"handleInformInit: received","streamId":"q9vcmje5","id":"1(@)"}
+{"time":"2026-08-14T21:31:54.062397411Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"q9vcmje5","id":"1(@)"}
+{"time":"2026-08-14T21:31:59.411924652Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"32s1adqk73rr"}
+{"time":"2026-08-14T21:45:48.396635188Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"32s1adqk73rr"}
+{"time":"2026-08-14T21:45:48.929834053Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"32s1adqk73rr"}
+{"time":"2026-08-14T21:45:48.931620539Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"q9vcmje5","id":"1(@)"}
+{"time":"2026-08-14T21:45:48.932491197Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"q9vcmje5","id":"1(@)"}
+{"time":"2026-08-14T21:45:49.464388506Z","level":"INFO","msg":"handleInformInit: received","streamId":"k24y90ws","id":"1(@)"}
+{"time":"2026-08-14T21:45:49.733853364Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"k24y90ws","id":"1(@)"}
+{"time":"2026-08-14T21:45:55.329792921Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"dr322rnt7xgx"}
+{"time":"2026-08-14T21:59:01.158537572Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"dr322rnt7xgx"}
+{"time":"2026-08-14T21:59:01.914569559Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"dr322rnt7xgx"}
+{"time":"2026-08-14T21:59:01.916156477Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"k24y90ws","id":"1(@)"}
+{"time":"2026-08-14T21:59:01.917534874Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"k24y90ws","id":"1(@)"}
+{"time":"2026-08-14T21:59:02.524868682Z","level":"INFO","msg":"handleInformInit: received","streamId":"npap5k3g","id":"1(@)"}
+{"time":"2026-08-14T21:59:02.786199875Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"npap5k3g","id":"1(@)"}
+{"time":"2026-08-14T21:59:08.192858634Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"2g99yy5pussw"}
+{"time":"2026-08-14T22:11:53.643379161Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"2g99yy5pussw"}
+{"time":"2026-08-14T22:11:54.173579992Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"2g99yy5pussw"}
+{"time":"2026-08-14T22:11:54.17490129Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"npap5k3g","id":"1(@)"}
+{"time":"2026-08-14T22:11:54.175479186Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"npap5k3g","id":"1(@)"}
+{"time":"2026-08-14T22:11:54.694802218Z","level":"INFO","msg":"handleInformInit: received","streamId":"a3607q94","id":"1(@)"}
+{"time":"2026-08-14T22:11:54.956642682Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"a3607q94","id":"1(@)"}
+{"time":"2026-08-14T22:12:00.441973218Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"raaqqqmajsfs"}
+{"time":"2026-08-14T22:24:55.662446346Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"raaqqqmajsfs"}
+{"time":"2026-08-14T22:24:56.296327526Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"raaqqqmajsfs"}
+{"time":"2026-08-14T22:24:56.297810539Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"a3607q94","id":"1(@)"}
+{"time":"2026-08-14T22:24:56.29856977Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"a3607q94","id":"1(@)"}
+{"time":"2026-08-14T22:24:56.82560196Z","level":"INFO","msg":"handleInformInit: received","streamId":"wcvy5vyw","id":"1(@)"}
+{"time":"2026-08-14T22:24:57.085772231Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"wcvy5vyw","id":"1(@)"}
+{"time":"2026-08-14T22:25:02.571913681Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"2nb8wusqu4fo"}
+{"time":"2026-08-14T22:38:28.612623637Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"2nb8wusqu4fo"}
+{"time":"2026-08-14T22:38:29.208727987Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"2nb8wusqu4fo"}
+{"time":"2026-08-14T22:38:29.210106262Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"wcvy5vyw","id":"1(@)"}
+{"time":"2026-08-14T22:38:29.210705504Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"wcvy5vyw","id":"1(@)"}
+{"time":"2026-08-14T22:38:29.839594174Z","level":"INFO","msg":"handleInformInit: received","streamId":"uuv4h936","id":"1(@)"}
+{"time":"2026-08-14T22:38:30.099444184Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"uuv4h936","id":"1(@)"}
+{"time":"2026-08-14T22:38:35.620666929Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"r0aouorh27ls"}
+{"time":"2026-08-14T22:52:17.704425638Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"r0aouorh27ls"}
+{"time":"2026-08-14T22:52:18.35448297Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"r0aouorh27ls"}
+{"time":"2026-08-14T22:52:18.355720686Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"uuv4h936","id":"1(@)"}
+{"time":"2026-08-14T22:52:18.356578583Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"uuv4h936","id":"1(@)"}
+{"time":"2026-08-14T22:52:18.899799881Z","level":"INFO","msg":"handleInformInit: received","streamId":"x87hwrtd","id":"1(@)"}
+{"time":"2026-08-14T22:52:19.159494071Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"x87hwrtd","id":"1(@)"}
+{"time":"2026-08-14T22:52:24.671083701Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"hn7lqjhi3ekb"}
+{"time":"2026-08-14T23:05:51.214192173Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"hn7lqjhi3ekb"}
+{"time":"2026-08-14T23:05:51.783107435Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"hn7lqjhi3ekb"}
+{"time":"2026-08-14T23:05:51.784410941Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"x87hwrtd","id":"1(@)"}
+{"time":"2026-08-14T23:05:51.78516667Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"x87hwrtd","id":"1(@)"}
+{"time":"2026-08-14T23:05:52.088192277Z","level":"INFO","msg":"handleInformTeardown: server teardown initiated","id":"1(@)"}
+{"time":"2026-08-14T23:05:52.088229648Z","level":"INFO","msg":"handleInformTeardown: server shutdown complete","id":"1(@)"}
+{"time":"2026-08-14T23:05:52.088239651Z","level":"INFO","msg":"connection: closing","id":"1(@)"}
+{"time":"2026-08-14T23:05:52.088243873Z","level":"INFO","msg":"server: is shutting down"}
+{"time":"2026-08-14T23:05:52.0882852Z","level":"INFO","msg":"connection: closed successfully","id":"1(@)"}
+{"time":"2026-08-14T23:05:52.088299095Z","level":"INFO","msg":"processOutgoingData: finished","id":"1(@)"}
+{"time":"2026-08-14T23:05:52.088345917Z","level":"INFO","msg":"connection: ManageConnectionData: connection closed","id":"1(@)"}
+{"time":"2026-08-14T23:05:52.088452051Z","level":"INFO","msg":"server: listener closed","addr":{"Name":"/tmp/wandb-1229423-1230766-2604858008/socket","Net":"unix"}}
+{"time":"2026-08-14T23:05:52.088520571Z","level":"INFO","msg":"server: all connections closed"}

zain/Activation/wandb/run-20260814_221154-a3607q94/logs/debug-internal.log

@@ -0,0 +1,121 @@
+{"time":"2026-08-14T22:11:54.694930488Z","level":"INFO","msg":"wandb-core"}
+{"time":"2026-08-14T22:11:54.695061314Z","level":"INFO","msg":"stream: starting","core version":"0.28.1"}
+{"time":"2026-08-14T22:11:54.956465986Z","level":"INFO","msg":"stream: created new stream","id":"a3607q94"}
+{"time":"2026-08-14T22:11:54.956541268Z","level":"INFO","msg":"handler: started"}
+{"time":"2026-08-14T22:11:54.95663725Z","level":"INFO","msg":"stream: started"}
+{"time":"2026-08-14T22:11:54.956647578Z","level":"INFO","msg":"writer: started","stream_id":"a3607q94"}
+{"time":"2026-08-14T22:11:54.9566745Z","level":"INFO","msg":"sender: started"}
+{"time":"2026-08-14T22:11:55.405026716Z","level":"INFO","msg":"filestream: sending request","total_files":1,"console_offset":0,"console_lines":1}
+{"time":"2026-08-14T22:11:55.518319573Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:12:10.405391964Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":0,"events_lines":1,"console_offset":0,"console_lines":1,"uploaded_len":2}
+{"time":"2026-08-14T22:12:10.559461859Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:12:25.405485607Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":0,"history_lines":1,"events_offset":1,"events_lines":2,"console_offset":0,"console_lines":2}
+{"time":"2026-08-14T22:12:25.550239937Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:12:40.405268298Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":1,"history_lines":1,"events_offset":3,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-14T22:12:40.5116701Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:12:55.406850003Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":2,"history_lines":1,"events_offset":5,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-14T22:12:55.555408593Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:13:10.405496619Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":3,"history_lines":1,"events_offset":7,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-14T22:13:10.5558936Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:13:25.405663028Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":4,"history_lines":1,"events_offset":9,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-14T22:13:25.538451856Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:13:40.405804156Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":5,"history_lines":1,"events_offset":11,"events_lines":2,"console_offset":2,"console_lines":10}
+{"time":"2026-08-14T22:13:40.533262745Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:13:55.405648498Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":6,"history_lines":1,"events_offset":13,"events_lines":2,"console_offset":10,"console_lines":1}
+{"time":"2026-08-14T22:13:55.529428811Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:14:10.405699023Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":7,"history_lines":1,"events_offset":15,"events_lines":2,"console_offset":10,"console_lines":1}
+{"time":"2026-08-14T22:14:10.554609811Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:14:25.405296636Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":8,"history_lines":1,"events_offset":17,"events_lines":2,"console_offset":10,"console_lines":1}
+{"time":"2026-08-14T22:14:25.544426629Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:14:40.405384171Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":9,"history_lines":1,"events_offset":19,"events_lines":2,"console_offset":10,"console_lines":1}
+{"time":"2026-08-14T22:14:40.53709818Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:14:55.405885358Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":10,"history_lines":1,"events_offset":21,"events_lines":2,"console_offset":12,"console_lines":10}
+{"time":"2026-08-14T22:14:55.538990888Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:15:10.405566821Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":11,"history_lines":1,"events_offset":23,"events_lines":2,"console_offset":20,"console_lines":1}
+{"time":"2026-08-14T22:15:10.550080507Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:15:25.405918015Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":12,"history_lines":1,"events_offset":25,"events_lines":2,"console_offset":20,"console_lines":1}
+{"time":"2026-08-14T22:15:25.513709253Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:15:40.405536965Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":13,"history_lines":1,"events_offset":27,"events_lines":2,"console_offset":20,"console_lines":1}
+{"time":"2026-08-14T22:15:40.505023493Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:15:55.405149988Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":14,"history_lines":1,"events_offset":29,"events_lines":2,"console_offset":20,"console_lines":1}
+{"time":"2026-08-14T22:15:55.51412857Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:16:10.405144989Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":15,"history_lines":1,"events_offset":31,"events_lines":2,"console_offset":22,"console_lines":10}
+{"time":"2026-08-14T22:16:10.541001666Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:16:25.405968509Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":16,"history_lines":1,"events_offset":33,"events_lines":2,"console_offset":30,"console_lines":1}
+{"time":"2026-08-14T22:16:25.508694054Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:16:40.405798516Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":17,"history_lines":1,"events_offset":35,"events_lines":2,"console_offset":30,"console_lines":1}
+{"time":"2026-08-14T22:16:40.533724449Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:16:55.405448681Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":18,"history_lines":1,"events_offset":37,"events_lines":2,"console_offset":30,"console_lines":1}
+{"time":"2026-08-14T22:16:55.528250746Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:17:10.406148823Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":19,"history_lines":1,"events_offset":39,"events_lines":2,"console_offset":30,"console_lines":1}
+{"time":"2026-08-14T22:17:10.549567524Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:17:25.405294454Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":20,"history_lines":1,"events_offset":41,"events_lines":2,"console_offset":32,"console_lines":10}
+{"time":"2026-08-14T22:17:25.532071927Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:17:40.405762467Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":21,"history_lines":1,"events_offset":43,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-14T22:17:40.528553533Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:17:55.405245047Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":22,"history_lines":1,"events_offset":45,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-14T22:17:55.535139755Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:18:10.405494472Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":23,"history_lines":1,"events_offset":47,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-14T22:18:10.556962534Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:18:25.405906254Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":24,"history_lines":1,"events_offset":49,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-14T22:18:25.543583377Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:18:40.405684146Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":25,"history_lines":1,"events_offset":51,"events_lines":2,"console_offset":42,"console_lines":10}
+{"time":"2026-08-14T22:18:40.517192829Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:18:55.405352723Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":26,"history_lines":1,"events_offset":53,"events_lines":2,"console_offset":50,"console_lines":1}
+{"time":"2026-08-14T22:18:55.556085442Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:19:10.405214712Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":27,"history_lines":1,"events_offset":55,"events_lines":2,"console_offset":50,"console_lines":1}
+{"time":"2026-08-14T22:19:10.549268441Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:19:25.405984196Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":28,"history_lines":1,"events_offset":57,"events_lines":2,"console_offset":50,"console_lines":1}
+{"time":"2026-08-14T22:19:25.524400916Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:19:40.406034794Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":29,"history_lines":1,"events_offset":59,"events_lines":2,"console_offset":50,"console_lines":1}
+{"time":"2026-08-14T22:19:40.529752452Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:19:55.405562656Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":30,"history_lines":1,"events_offset":61,"events_lines":2,"console_offset":52,"console_lines":10}
+{"time":"2026-08-14T22:19:55.522207405Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:20:10.405470142Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":31,"history_lines":1,"events_offset":63,"events_lines":2,"console_offset":60,"console_lines":1}
+{"time":"2026-08-14T22:20:10.520670867Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:20:25.40533656Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":32,"history_lines":1,"events_offset":65,"events_lines":2,"console_offset":60,"console_lines":1}
+{"time":"2026-08-14T22:20:25.548953084Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:20:40.405524309Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":33,"history_lines":1,"events_offset":67,"events_lines":2,"console_offset":60,"console_lines":1}
+{"time":"2026-08-14T22:20:40.546057054Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:20:55.405924314Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":34,"history_lines":1,"events_offset":69,"events_lines":2,"console_offset":60,"console_lines":1}
+{"time":"2026-08-14T22:20:55.51483369Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:21:10.405655258Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":35,"history_lines":1,"events_offset":71,"events_lines":2,"console_offset":62,"console_lines":10}
+{"time":"2026-08-14T22:21:10.538990403Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:21:25.405590498Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":36,"history_lines":1,"events_offset":73,"events_lines":2,"console_offset":70,"console_lines":1}
+{"time":"2026-08-14T22:21:25.560039008Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:21:40.406004609Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":37,"history_lines":1,"events_offset":75,"events_lines":2,"console_offset":70,"console_lines":1}
+{"time":"2026-08-14T22:21:40.520368533Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:21:55.40554939Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":38,"history_lines":1,"events_offset":77,"events_lines":2,"console_offset":70,"console_lines":1}
+{"time":"2026-08-14T22:21:55.510133609Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:22:10.406035723Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":39,"history_lines":1,"events_offset":79,"events_lines":2,"console_offset":70,"console_lines":1}
+{"time":"2026-08-14T22:22:10.564369093Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:22:25.405797515Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":40,"history_lines":1,"events_offset":81,"events_lines":2,"console_offset":72,"console_lines":10}
+{"time":"2026-08-14T22:22:25.518466895Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:22:40.40573414Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":41,"history_lines":1,"events_offset":83,"events_lines":2,"console_offset":80,"console_lines":1}
+{"time":"2026-08-14T22:22:40.565407197Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:22:55.405279582Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":42,"history_lines":1,"events_offset":85,"events_lines":2,"console_offset":80,"console_lines":1}
+{"time":"2026-08-14T22:22:55.572896644Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:23:10.406123059Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":43,"history_lines":1,"events_offset":87,"events_lines":2,"console_offset":80,"console_lines":1}
+{"time":"2026-08-14T22:23:10.525921175Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:23:25.405432496Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":44,"history_lines":1,"events_offset":89,"events_lines":2,"console_offset":80,"console_lines":1}
+{"time":"2026-08-14T22:23:25.587919865Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:23:40.405843481Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":45,"history_lines":1,"events_offset":91,"events_lines":2,"console_offset":82,"console_lines":10}
+{"time":"2026-08-14T22:23:40.555604866Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:23:55.405877105Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":46,"history_lines":1,"events_offset":93,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-14T22:23:55.513658618Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:24:10.405810362Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":47,"history_lines":1,"events_offset":95,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-14T22:24:10.565224461Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:24:25.406151072Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":48,"history_lines":1,"events_offset":97,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-14T22:24:25.552886878Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:24:40.405160685Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":49,"history_lines":1,"events_offset":99,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-14T22:24:40.52176699Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:24:55.406073053Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":50,"history_lines":2,"events_offset":101,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-14T22:24:55.539961111Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:24:56.042584292Z","level":"INFO","msg":"fileTransfer: Close: file transfer manager closed"}
+{"time":"2026-08-14T22:24:56.042849717Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":52,"history_lines":1,"events_offset":103,"events_lines":1,"console_offset":92,"console_lines":18,"uploaded_len":3,"complete":true,"exit_code":0}
+{"time":"2026-08-14T22:24:56.293989988Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:24:56.295208855Z","level":"INFO","msg":"handler: operation stats","stats":{}}
+{"time":"2026-08-14T22:24:56.297829968Z","level":"INFO","msg":"stream: finishing up"}
+{"time":"2026-08-14T22:24:56.297843095Z","level":"INFO","msg":"handler: closed"}
+{"time":"2026-08-14T22:24:56.297914854Z","level":"INFO","msg":"sender: closed"}
+{"time":"2026-08-14T22:24:56.297920806Z","level":"INFO","msg":"stream: all finished"}

zain/Activation/wandb/run-20260814_221154-a3607q94/logs/debug.log

@@ -0,0 +1,24 @@
+2026-08-14 22:11:54,693 INFO    MainThread:1229423 [wandb_init.py:setup_run_log_directory():729] Logging user logs to /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/wandb/run-20260814_221154-a3607q94/logs/debug.log
+2026-08-14 22:11:54,693 INFO    MainThread:1229423 [wandb_init.py:setup_run_log_directory():730] Logging internal logs to /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/wandb/run-20260814_221154-a3607q94/logs/debug-internal.log
+2026-08-14 22:11:54,693 INFO    MainThread:1229423 [wandb_init.py:init():772] calling init triggers
+2026-08-14 22:11:54,693 INFO    MainThread:1229423 [wandb_init.py:init():777] wandb.init called with sweep_config: {}
+config: {'_wandb': {}}
+2026-08-14 22:11:54,693 INFO    MainThread:1229423 [wandb_init.py:init():820] starting backend
+2026-08-14 22:11:54,693 INFO    MainThread:1229423 [wandb_init.py:init():835] sending inform_init request
+2026-08-14 22:11:54,957 INFO    MainThread:1229423 [wandb_init.py:init():840] backend started and connected
+2026-08-14 22:11:54,958 INFO    MainThread:1229423 [wandb_init.py:init():910] updated telemetry
+2026-08-14 22:11:54,965 INFO    MainThread:1229423 [wandb_init.py:init():933] communicating run to backend with 90.0 second timeout
+2026-08-14 22:11:55,322 INFO    MainThread:1229423 [wandb_init.py:init():978] starting run threads in backend
+2026-08-14 22:11:55,399 INFO    MainThread:1229423 [wandb_run.py:_console_start():2621] atexit reg
+2026-08-14 22:11:55,399 INFO    MainThread:1229423 [wandb_run.py:_redirect():2471] redirect: wrap_raw
+2026-08-14 22:11:55,399 INFO    MainThread:1229423 [wandb_run.py:_redirect():2540] Wrapping output streams.
+2026-08-14 22:11:55,399 INFO    MainThread:1229423 [wandb_run.py:_redirect():2563] Redirects installed.
+2026-08-14 22:11:55,400 INFO    MainThread:1229423 [wandb_init.py:init():1016] run started, returning control to user process
+2026-08-14 22:11:55,401 INFO    MainThread:1229423 [wandb_run.py:_config_callback():1346] config_cb None None {'transformers_version': '5.16.0.dev0', 'architectures': None, 'output_hidden_states': False, 'return_dict': True, 'dtype': None, 'chunk_size_feed_forward': 0, 'is_encoder_decoder': False, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'problem_type': None, 'vocab_size': 4096, 'hidden_size': 128, 'intermediate_size': 256, 'num_hidden_layers': 21, 'num_attention_heads': 4, 'num_key_value_heads': 4, 'hidden_act': 'silu', 'max_position_embeddings': 512, 'initializer_range': 0.02, 'rms_norm_eps': 1e-06, 'use_cache': False, 'pad_token_id': 0, 'bos_token_id': 1, 'eos_token_id': 2, 'pretraining_tp': 1, 'tie_word_embeddings': True, 'rope_parameters': {'rope_theta': 10000.0, 'rope_type': 'default'}, 'attention_bias': False, 'attention_dropout': 0.0, 'mlp_bias': False, 'head_dim': 32, '_name_or_path': '', 'tokenizer_name': 'w-ahmad/tiny-stories-tokenizer', 'mlp_type': 'mlp', 'activation': 'silu-waleed10', 'waleed_beta': 10.0, 'model_type': 'tiny_llama', 'output_attentions': False, 'output_dir': 'out/mlp-silu-waleed10-21L_run', 'per_device_train_batch_size': 128, 'num_train_epochs': 1, 'max_steps': 1000, 'learning_rate': 0.001, 'lr_scheduler_type': 'constant', 'lr_scheduler_kwargs': None, 'warmup_steps': 0, 'optim': 'adamw_torch', 'optim_args': None, 'weight_decay': 0.01, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'optim_target_modules': None, 'gradient_accumulation_steps': 4, 'average_tokens_across_devices': True, 'max_grad_norm': 1.0, 'label_smoothing_factor': 0.0, 'bf16': True, 'fp16': False, 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'gradient_checkpointing': False, 'gradient_checkpointing_kwargs': None, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'use_liger_kernel': False, 'liger_kernel_config': None, 'neftune_noise_alpha': None, 'torch_empty_cache_steps': None, 'auto_find_batch_size': False, 'logging_strategy': 'steps', 'logging_steps': 20, 'logging_first_step': False, 'log_on_each_node': True, 'logging_nan_inf_filter': True, 'include_num_input_tokens_seen': 'no', 'log_level': 'passive', 'log_level_replica': 'warning', 'disable_tqdm': False, 'report_to': ['wandb'], 'run_name': 'LM-mlp-silu-waleed10-21L-4.0M-20260814-221154', 'project': 'huggingface', 'trackio_space_id': None, 'trackio_bucket_id': None, 'trackio_static_space_id': None, 'eval_strategy': 'steps', 'eval_steps': 60000, 'eval_delay': 0, 'per_device_eval_batch_size': 128, 'prediction_loss_only': False, 'eval_on_start': False, 'eval_do_concat_batches': True, 'eval_use_gather_object': False, 'eval_accumulation_steps': None, 'include_for_metrics': [], 'batch_eval_metrics': False, 'save_only_model': False, 'save_strategy': 'steps', 'save_steps': 100, 'save_on_each_node': False, 'save_total_limit': None, 'enable_jit_checkpoint': False, 'push_to_hub': False, 'hub_token': '<HUB_TOKEN>', 'hub_private_repo': None, 'hub_model_id': 'w-ahmad/A-mlp-silu-waleed10-21L', 'hub_strategy': 'every_save', 'hub_always_push': False, 'hub_revision': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'restore_callback_states_from_checkpoint': False, 'full_determinism': False, 'seed': 42, 'data_seed': 42, 'use_cpu': False, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}, 'parallelism_config': None, 'dataloader_drop_last': False, 'dataloader_num_workers': 0, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'dataloader_prefetch_factor': None, 'dataloader_multiprocessing_context': None, 'dataloader_in_order': True, 'remove_unused_columns': False, 'label_names': None, 'train_sampling_strategy': 'random', 'length_column_name': 'length', 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'ddp_static_graph': None, 'ddp_backend': None, 'ddp_timeout': 1800, 'fsdp': None, 'fsdp_config': None, 'deepspeed': None, 'debug': [], 'skip_memory_metrics': True, 'do_train': False, 'do_eval': True, 'do_predict': False, 'resume_from_checkpoint': None, 'local_rank': -1}
+2026-08-14 22:11:55,404 INFO    MainThread:1229423 [wandb_config.py:__setitem__():155] [no run ID] config set model/num_parameters = 3970432 - <bound method Run._config_callback of <wandb.sdk.wandb_run.Run object at 0x1528dffbec10>>
+2026-08-14 22:11:55,404 INFO    MainThread:1229423 [wandb_run.py:_config_callback():1346] config_cb model/num_parameters 3970432 None
+2026-08-14 22:24:55,661 INFO    MainThread:1229423 [wandb_run.py:_finish():2383] finishing run deepnevro-deepnevro/research1/a3607q94
+2026-08-14 22:24:55,662 INFO    MainThread:1229423 [wandb_run.py:_atexit_cleanup():2588] got exitcode: 0
+2026-08-14 22:24:55,662 INFO    MainThread:1229423 [wandb_run.py:_restore():2570] restore
+2026-08-14 22:24:55,662 INFO    MainThread:1229423 [wandb_run.py:_restore():2576] restore done
+2026-08-14 22:24:56,297 INFO    MainThread:1229423 [wandb_run.py:_footer_sync_info():3993] logging synced files

zain/Activation/wandb/run-20260814_221154-a3607q94/run-a3607q94.wandb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd675d3647f508d189770d90b8b24f154361d7cb4eb0a260d80b438c86d0d42b
+size 574747

zain/Activation/wandb/run-20260814_222456-wcvy5vyw/files/config.yaml

@@ -0,0 +1,439 @@
+_name_or_path:
+    value: ""
+_wandb:
+    value:
+        cli_version: 0.28.1
+        e:
+            ry559cdhs8pu8maqeo36io5wxu2p8d4g:
+                args:
+                    - --config
+                    - /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/configs/baseline.yaml
+                    - --variants
+                    - glu-waleed10
+                    - glu-silu-waleed10
+                    - mlp-waleed10
+                    - mlp-silu-waleed10
+                    - glu-waleed
+                    - glu-situglu_low
+                    - glu-waleedglu_low
+                codePath: sweep.py
+                codePathLocal: sweep.py
+                cpu_count: 112
+                cpu_count_logical: 224
+                cudaVersion: "12.4"
+                disk:
+                    /:
+                        total: "1560765693952"
+                        used: "709103636480"
+                email: deepnevro@gmail.com
+                executable: /mnt/data/zainulabideen/zain-exp/notebooks/my_env/bin/python
+                git:
+                    commit: 3fb47c6943d6927eafc39bd399a3d3a520bae74a
+                    remote: https://github.com/w-ahmad1a10/Activation.git
+                gpu: NVIDIA H100 80GB HBM3
+                gpu_count: 8
+                gpu_nvidia:
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-39c684a5-fde6-83d7-1663-0859795881ae
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-68012e5a-38b6-b643-0ca6-62fb66720bf3
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-132944c4-b689-2b5f-89a4-d730401677ab
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-2df386cc-6d26-d0e2-7a2d-a057b0d95864
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-bfa16575-1d94-1aa2-4537-2c93433f42ef
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-bc6c3e3c-9b90-09ca-c034-774961847c54
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-00a441e1-7c95-e7d6-4c35-43d6b291aea9
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-1c4d29a2-4647-6fce-d8fc-0c5ecfbbd6ea
+                host: deeplens-k3s-node1
+                memory:
+                    total: "2164089937920"
+                os: Linux-5.15.0-126-generic-x86_64-with-glibc2.35
+                program: /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/sweep.py
+                python: CPython 3.11.15
+                root: /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation
+                startedAt: "2026-08-14T22:24:56.823342Z"
+                writerId: ry559cdhs8pu8maqeo36io5wxu2p8d4g
+        m:
+            - "1": train/global_step
+              "6":
+                - 3
+              "7": []
+            - "2": '*'
+              "5": 1
+              "6":
+                - 1
+              "7": []
+        python_version: 3.11.15
+        t:
+            "1":
+                - 1
+                - 5
+                - 11
+                - 41
+                - 49
+                - 51
+                - 53
+                - 71
+            "2":
+                - 1
+                - 5
+                - 11
+                - 41
+                - 49
+                - 51
+                - 53
+                - 71
+            "3":
+                - 2
+                - 7
+                - 13
+                - 19
+                - 62
+                - 66
+            "4": 3.11.15
+            "5": 0.28.1
+            "6": 5.16.0.dev0
+            "9":
+                "1": transformers_trainer
+            "12": 0.28.1
+            "13": linux-x86_64
+accelerator_config:
+    value:
+        dispatch_batches: null
+        even_batches: true
+        gradient_accumulation_kwargs: null
+        non_blocking: false
+        split_batches: false
+        use_seedable_sampler: true
+activation:
+    value: waleed
+adam_beta1:
+    value: 0.9
+adam_beta2:
+    value: 0.999
+adam_epsilon:
+    value: 1e-08
+architectures:
+    value: null
+attention_bias:
+    value: false
+attention_dropout:
+    value: 0
+auto_find_batch_size:
+    value: false
+average_tokens_across_devices:
+    value: true
+batch_eval_metrics:
+    value: false
+bf16:
+    value: true
+bf16_full_eval:
+    value: false
+bos_token_id:
+    value: 1
+chunk_size_feed_forward:
+    value: 0
+data_seed:
+    value: 42
+dataloader_drop_last:
+    value: false
+dataloader_in_order:
+    value: true
+dataloader_multiprocessing_context:
+    value: null
+dataloader_num_workers:
+    value: 0
+dataloader_persistent_workers:
+    value: false
+dataloader_pin_memory:
+    value: true
+dataloader_prefetch_factor:
+    value: null
+ddp_backend:
+    value: null
+ddp_broadcast_buffers:
+    value: null
+ddp_bucket_cap_mb:
+    value: null
+ddp_find_unused_parameters:
+    value: null
+ddp_static_graph:
+    value: null
+ddp_timeout:
+    value: 1800
+debug:
+    value: []
+deepspeed:
+    value: null
+disable_tqdm:
+    value: false
+do_eval:
+    value: true
+do_predict:
+    value: false
+do_train:
+    value: false
+dtype:
+    value: null
+enable_jit_checkpoint:
+    value: false
+eos_token_id:
+    value: 2
+eval_accumulation_steps:
+    value: null
+eval_delay:
+    value: 0
+eval_do_concat_batches:
+    value: true
+eval_on_start:
+    value: false
+eval_steps:
+    value: 60000
+eval_strategy:
+    value: steps
+eval_use_gather_object:
+    value: false
+fp16:
+    value: false
+fp16_full_eval:
+    value: false
+fsdp:
+    value: null
+fsdp_config:
+    value: null
+full_determinism:
+    value: false
+gradient_accumulation_steps:
+    value: 4
+gradient_checkpointing:
+    value: false
+gradient_checkpointing_kwargs:
+    value: null
+greater_is_better:
+    value: null
+head_dim:
+    value: 32
+hidden_act:
+    value: silu
+hidden_size:
+    value: 128
+hub_always_push:
+    value: false
+hub_model_id:
+    value: w-ahmad/A-glu-waleed-21L
+hub_private_repo:
+    value: null
+hub_revision:
+    value: null
+hub_strategy:
+    value: every_save
+hub_token:
+    value: <HUB_TOKEN>
+id2label:
+    value:
+        "0": LABEL_0
+        "1": LABEL_1
+ignore_data_skip:
+    value: false
+include_for_metrics:
+    value: []
+include_num_input_tokens_seen:
+    value: "no"
+initializer_range:
+    value: 0.02
+intermediate_size:
+    value: 256
+is_encoder_decoder:
+    value: false
+label_names:
+    value: null
+label_smoothing_factor:
+    value: 0
+label2id:
+    value:
+        LABEL_0: 0
+        LABEL_1: 1
+learning_rate:
+    value: 0.001
+length_column_name:
+    value: length
+liger_kernel_config:
+    value: null
+load_best_model_at_end:
+    value: false
+local_rank:
+    value: -1
+log_level:
+    value: passive
+log_level_replica:
+    value: warning
+log_on_each_node:
+    value: true
+logging_first_step:
+    value: false
+logging_nan_inf_filter:
+    value: true
+logging_steps:
+    value: 20
+logging_strategy:
+    value: steps
+lr_scheduler_kwargs:
+    value: null
+lr_scheduler_type:
+    value: constant
+max_grad_norm:
+    value: 1
+max_position_embeddings:
+    value: 512
+max_steps:
+    value: 1000
+metric_for_best_model:
+    value: null
+mlp_bias:
+    value: false
+mlp_type:
+    value: glu
+model/num_parameters:
+    value: 3970432
+model_type:
+    value: tiny_llama
+neftune_noise_alpha:
+    value: null
+num_attention_heads:
+    value: 4
+num_hidden_layers:
+    value: 21
+num_key_value_heads:
+    value: 4
+num_train_epochs:
+    value: 1
+optim:
+    value: adamw_torch
+optim_args:
+    value: null
+optim_target_modules:
+    value: null
+output_attentions:
+    value: false
+output_dir:
+    value: out/glu-waleed-21L_run
+output_hidden_states:
+    value: false
+pad_token_id:
+    value: 0
+parallelism_config:
+    value: null
+per_device_eval_batch_size:
+    value: 128
+per_device_train_batch_size:
+    value: 128
+prediction_loss_only:
+    value: false
+pretraining_tp:
+    value: 1
+problem_type:
+    value: null
+project:
+    value: huggingface
+push_to_hub:
+    value: false
+remove_unused_columns:
+    value: false
+report_to:
+    value:
+        - wandb
+restore_callback_states_from_checkpoint:
+    value: false
+resume_from_checkpoint:
+    value: null
+return_dict:
+    value: true
+rms_norm_eps:
+    value: 1e-06
+rope_parameters:
+    value:
+        rope_theta: 10000
+        rope_type: default
+run_name:
+    value: LM-glu-waleed-21L-4.0M-20260814-222456
+save_on_each_node:
+    value: false
+save_only_model:
+    value: false
+save_steps:
+    value: 100
+save_strategy:
+    value: steps
+save_total_limit:
+    value: null
+seed:
+    value: 42
+skip_memory_metrics:
+    value: true
+tf32:
+    value: null
+tie_word_embeddings:
+    value: true
+tokenizer_name:
+    value: w-ahmad/tiny-stories-tokenizer
+torch_compile:
+    value: false
+torch_compile_backend:
+    value: null
+torch_compile_mode:
+    value: null
+torch_empty_cache_steps:
+    value: null
+trackio_bucket_id:
+    value: null
+trackio_space_id:
+    value: null
+trackio_static_space_id:
+    value: null
+train_sampling_strategy:
+    value: random
+transformers_version:
+    value: 5.16.0.dev0
+use_cache:
+    value: false
+use_cpu:
+    value: false
+use_liger_kernel:
+    value: false
+vocab_size:
+    value: 4096
+waleed_beta:
+    value: 10
+warmup_steps:
+    value: 0
+weight_decay:
+    value: 0.01

zain/Activation/wandb/run-20260814_222456-wcvy5vyw/files/output.log

@@ -0,0 +1,110 @@
+ 10%|█         | 100/1000 [01:19<11:47,  1.27it/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': '28.88', 'grad_norm': '4', 'learning_rate': '0.001', 'epoch': '0.01078', 'train/total_time_seconds': '8.163', 'train/time_per_step_avg': '0.4082', 'train/epoch_time_elapsed': '15.97', 'train/estimated_remaining_minutes': '6.667'}
+{'loss': '24.01', 'grad_norm': '1.969', 'learning_rate': '0.001', 'epoch': '0.02157', 'train/total_time_seconds': '16.29', 'train/time_per_step_avg': '0.4071', 'train/epoch_time_elapsed': '31.88', 'train/estimated_remaining_minutes': '6.514'}
+{'loss': '22.25', 'grad_norm': '1.992', 'learning_rate': '0.001', 'epoch': '0.03235', 'train/total_time_seconds': '24.42', 'train/time_per_step_avg': '0.407', 'train/epoch_time_elapsed': '47.66', 'train/estimated_remaining_minutes': '6.376'}
+{'loss': '20.39', 'grad_norm': '1.625', 'learning_rate': '0.001', 'epoch': '0.04314', 'train/total_time_seconds': '32.55', 'train/time_per_step_avg': '0.4069', 'train/epoch_time_elapsed': '63.47', 'train/estimated_remaining_minutes': '6.239'}
+{'loss': '18.54', 'grad_norm': '2.062', 'learning_rate': '0.001', 'epoch': '0.05392', 'train/total_time_seconds': '40.67', 'train/time_per_step_avg': '0.4067', 'train/epoch_time_elapsed': '79.22', 'train/estimated_remaining_minutes': '6.1'}
+  - 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, 75.75it/s]
+ 20%|██        | 200/1000 [02:38<10:30,  1.27it/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': '17.23', 'grad_norm': '2.094', 'learning_rate': '0.001', 'epoch': '0.06471', 'train/total_time_seconds': '48.8', 'train/time_per_step_avg': '0.4064', 'train/epoch_time_elapsed': '95.07', 'train/estimated_remaining_minutes': '5.964'}
+{'loss': '16.2', 'grad_norm': '2.938', 'learning_rate': '0.001', 'epoch': '0.07549', 'train/total_time_seconds': '56.96', 'train/time_per_step_avg': '0.4067', 'train/epoch_time_elapsed': '110.9', 'train/estimated_remaining_minutes': '5.831'}
+{'loss': '15.42', 'grad_norm': '1.352', 'learning_rate': '0.001', 'epoch': '0.08628', 'train/total_time_seconds': '65.09', 'train/time_per_step_avg': '0.4067', 'train/epoch_time_elapsed': '126.6', 'train/estimated_remaining_minutes': '5.695'}
+{'loss': '14.78', 'grad_norm': '2.297', 'learning_rate': '0.001', 'epoch': '0.09706', 'train/total_time_seconds': '73.21', 'train/time_per_step_avg': '0.4065', 'train/epoch_time_elapsed': '142.4', 'train/estimated_remaining_minutes': '5.558'}
+{'loss': '14.2', 'grad_norm': '2.234', 'learning_rate': '0.001', 'epoch': '0.1078', 'train/total_time_seconds': '81.33', 'train/time_per_step_avg': '0.4066', 'train/epoch_time_elapsed': '158.2', 'train/estimated_remaining_minutes': '5.422'}
+  - 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, 76.92it/s]
+ 30%|███       | 300/1000 [03:57<09:04,  1.29it/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': '13.72', 'grad_norm': '2.484', 'learning_rate': '0.001', 'epoch': '0.1186', 'train/total_time_seconds': '89.45', 'train/time_per_step_avg': '0.4066', 'train/epoch_time_elapsed': '174.1', 'train/estimated_remaining_minutes': '5.286'}
+{'loss': '13.31', 'grad_norm': '1.75', 'learning_rate': '0.001', 'epoch': '0.1294', 'train/total_time_seconds': '97.59', 'train/time_per_step_avg': '0.4063', 'train/epoch_time_elapsed': '189.9', 'train/estimated_remaining_minutes': '5.151'}
+{'loss': '12.89', 'grad_norm': '2.547', 'learning_rate': '0.001', 'epoch': '0.1402', 'train/total_time_seconds': '105.7', 'train/time_per_step_avg': '0.4064', 'train/epoch_time_elapsed': '205.8', 'train/estimated_remaining_minutes': '5.015'}
+{'loss': '12.52', 'grad_norm': '2.438', 'learning_rate': '0.001', 'epoch': '0.151', 'train/total_time_seconds': '113.8', 'train/time_per_step_avg': '0.4064', 'train/epoch_time_elapsed': '221.6', 'train/estimated_remaining_minutes': '4.879'}
+{'loss': '12.18', 'grad_norm': '2.547', 'learning_rate': '0.001', 'epoch': '0.1618', 'train/total_time_seconds': '121.9', 'train/time_per_step_avg': '0.4062', 'train/epoch_time_elapsed': '237.4', 'train/estimated_remaining_minutes': '4.742'}
+  - 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, 79.43it/s]
+ 40%|████      | 400/1000 [05:16<07:52,  1.27it/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': '11.86', 'grad_norm': '2.094', 'learning_rate': '0.001', 'epoch': '0.1726', 'train/total_time_seconds': '130.1', 'train/time_per_step_avg': '0.406', 'train/epoch_time_elapsed': '253.1', 'train/estimated_remaining_minutes': '4.606'}
+{'loss': '11.58', 'grad_norm': '1.992', 'learning_rate': '0.001', 'epoch': '0.1833', 'train/total_time_seconds': '138.1', 'train/time_per_step_avg': '0.4055', 'train/epoch_time_elapsed': '268.9', 'train/estimated_remaining_minutes': '4.469'}
+{'loss': '11.32', 'grad_norm': '2.188', 'learning_rate': '0.001', 'epoch': '0.1941', 'train/total_time_seconds': '146.2', 'train/time_per_step_avg': '0.4051', 'train/epoch_time_elapsed': '284.6', 'train/estimated_remaining_minutes': '4.333'}
+{'loss': '11.07', 'grad_norm': '2.219', 'learning_rate': '0.001', 'epoch': '0.2049', 'train/total_time_seconds': '154.4', 'train/time_per_step_avg': '0.4051', 'train/epoch_time_elapsed': '300.4', 'train/estimated_remaining_minutes': '4.198'}
+{'loss': '10.86', 'grad_norm': '2.234', 'learning_rate': '0.001', 'epoch': '0.2157', 'train/total_time_seconds': '162.5', 'train/time_per_step_avg': '0.4053', 'train/epoch_time_elapsed': '316.2', 'train/estimated_remaining_minutes': '4.062'}
+  - 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, 74.34it/s]
+ 50%|█████     | 500/1000 [06:35<06:30,  1.28it/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': '10.65', 'grad_norm': '2.516', 'learning_rate': '0.001', 'epoch': '0.2265', 'train/total_time_seconds': '170.6', 'train/time_per_step_avg': '0.4052', 'train/epoch_time_elapsed': '332.1', 'train/estimated_remaining_minutes': '3.926'}
+{'loss': '10.53', 'grad_norm': '2.656', 'learning_rate': '0.001', 'epoch': '0.2373', 'train/total_time_seconds': '178.7', 'train/time_per_step_avg': '0.4053', 'train/epoch_time_elapsed': '347.9', 'train/estimated_remaining_minutes': '3.79'}
+{'loss': '10.34', 'grad_norm': '2.016', 'learning_rate': '0.001', 'epoch': '0.248', 'train/total_time_seconds': '186.8', 'train/time_per_step_avg': '0.4055', 'train/epoch_time_elapsed': '363.8', 'train/estimated_remaining_minutes': '3.655'}
+{'loss': '10.18', 'grad_norm': '2.281', 'learning_rate': '0.001', 'epoch': '0.2588', 'train/total_time_seconds': '194.9', 'train/time_per_step_avg': '0.4056', 'train/epoch_time_elapsed': '379.7', 'train/estimated_remaining_minutes': '3.519'}
+{'loss': '10.04', 'grad_norm': '2.422', 'learning_rate': '0.001', 'epoch': '0.2696', 'train/total_time_seconds': '203.1', 'train/time_per_step_avg': '0.4058', 'train/epoch_time_elapsed': '395.4', 'train/estimated_remaining_minutes': '3.384'}
+  - 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, 41.75it/s]
+ 60%|██████    | 600/1000 [07:54<05:19,  1.25it/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': '9.927', 'grad_norm': '1.977', 'learning_rate': '0.001', 'epoch': '0.2804', 'train/total_time_seconds': '211.2', 'train/time_per_step_avg': '0.4062', 'train/epoch_time_elapsed': '411.3', 'train/estimated_remaining_minutes': '3.249'}
+{'loss': '9.82', 'grad_norm': '2.172', 'learning_rate': '0.001', 'epoch': '0.2912', 'train/total_time_seconds': '219.3', 'train/time_per_step_avg': '0.4066', 'train/epoch_time_elapsed': '427.1', 'train/estimated_remaining_minutes': '3.114'}
+{'loss': '9.708', 'grad_norm': '2.594', 'learning_rate': '0.001', 'epoch': '0.302', 'train/total_time_seconds': '227.5', 'train/time_per_step_avg': '0.4066', 'train/epoch_time_elapsed': '442.8', 'train/estimated_remaining_minutes': '2.979'}
+{'loss': '9.597', 'grad_norm': '2.109', 'learning_rate': '0.001', 'epoch': '0.3128', 'train/total_time_seconds': '235.5', 'train/time_per_step_avg': '0.4063', 'train/epoch_time_elapsed': '458.5', 'train/estimated_remaining_minutes': '2.843'}
+{'loss': '9.496', 'grad_norm': '2.547', 'learning_rate': '0.001', 'epoch': '0.3235', 'train/total_time_seconds': '243.6', 'train/time_per_step_avg': '0.4059', 'train/epoch_time_elapsed': '474.5', 'train/estimated_remaining_minutes': '2.707'}
+  - 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, 80.61it/s]
+ 70%|███████   | 700/1000 [09:13<03:55,  1.27it/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': '9.41', 'grad_norm': '2.219', 'learning_rate': '0.001', 'epoch': '0.3343', 'train/total_time_seconds': '251.7', 'train/time_per_step_avg': '0.4053', 'train/epoch_time_elapsed': '490.4', 'train/estimated_remaining_minutes': '2.571'}
+{'loss': '9.316', 'grad_norm': '2.312', 'learning_rate': '0.001', 'epoch': '0.3451', 'train/total_time_seconds': '259.8', 'train/time_per_step_avg': '0.405', 'train/epoch_time_elapsed': '506.1', 'train/estimated_remaining_minutes': '2.436'}
+{'loss': '9.233', 'grad_norm': '1.945', 'learning_rate': '0.001', 'epoch': '0.3559', 'train/total_time_seconds': '268', 'train/time_per_step_avg': '0.4051', 'train/epoch_time_elapsed': '522', 'train/estimated_remaining_minutes': '2.301'}
+{'loss': '9.158', 'grad_norm': '2.125', 'learning_rate': '0.001', 'epoch': '0.3667', 'train/total_time_seconds': '276.1', 'train/time_per_step_avg': '0.4054', 'train/epoch_time_elapsed': '537.9', 'train/estimated_remaining_minutes': '2.165'}
+{'loss': '9.085', 'grad_norm': '2.094', 'learning_rate': '0.001', 'epoch': '0.3775', 'train/total_time_seconds': '284.2', 'train/time_per_step_avg': '0.4058', 'train/epoch_time_elapsed': '553.8', 'train/estimated_remaining_minutes': '2.03'}
+  - 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, 78.59it/s]
+ 80%|████████  | 800/1000 [10:33<02:41,  1.24it/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': '8.995', 'grad_norm': '1.977', 'learning_rate': '0.001', 'epoch': '0.3882', 'train/total_time_seconds': '292.4', 'train/time_per_step_avg': '0.4063', 'train/epoch_time_elapsed': '569.8', 'train/estimated_remaining_minutes': '1.895'}
+{'loss': '8.963', 'grad_norm': '2', 'learning_rate': '0.001', 'epoch': '0.399', 'train/total_time_seconds': '300.5', 'train/time_per_step_avg': '0.4063', 'train/epoch_time_elapsed': '585.6', 'train/estimated_remaining_minutes': '1.759'}
+{'loss': '8.899', 'grad_norm': '1.867', 'learning_rate': '0.001', 'epoch': '0.4098', 'train/total_time_seconds': '308.5', 'train/time_per_step_avg': '0.4059', 'train/epoch_time_elapsed': '601.6', 'train/estimated_remaining_minutes': '1.624'}
+{'loss': '8.858', 'grad_norm': '1.891', 'learning_rate': '0.001', 'epoch': '0.4206', 'train/total_time_seconds': '316.6', 'train/time_per_step_avg': '0.4056', 'train/epoch_time_elapsed': '617.4', 'train/estimated_remaining_minutes': '1.489'}
+{'loss': '8.76', 'grad_norm': '2.109', 'learning_rate': '0.001', 'epoch': '0.4314', 'train/total_time_seconds': '324.7', 'train/time_per_step_avg': '0.4052', 'train/epoch_time_elapsed': '633.3', 'train/estimated_remaining_minutes': '1.353'}
+  - 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, 79.93it/s]
+ 90%|█████████ | 900/1000 [11:52<01:19,  1.26it/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': '8.735', 'grad_norm': '2.594', 'learning_rate': '0.001', 'epoch': '0.4422', 'train/total_time_seconds': '332.8', 'train/time_per_step_avg': '0.4048', 'train/epoch_time_elapsed': '649.1', 'train/estimated_remaining_minutes': '1.218'}
+{'loss': '8.695', 'grad_norm': '2.25', 'learning_rate': '0.001', 'epoch': '0.453', 'train/total_time_seconds': '340.9', 'train/time_per_step_avg': '0.4049', 'train/epoch_time_elapsed': '664.9', 'train/estimated_remaining_minutes': '1.082'}
+{'loss': '8.626', 'grad_norm': '1.992', 'learning_rate': '0.001', 'epoch': '0.4637', 'train/total_time_seconds': '349.1', 'train/time_per_step_avg': '0.4051', 'train/epoch_time_elapsed': '680.6', 'train/estimated_remaining_minutes': '0.947'}
+{'loss': '8.586', 'grad_norm': '1.922', 'learning_rate': '0.001', 'epoch': '0.4745', 'train/total_time_seconds': '357.2', 'train/time_per_step_avg': '0.4056', 'train/epoch_time_elapsed': '696.5', 'train/estimated_remaining_minutes': '0.8118'}
+{'loss': '8.538', 'grad_norm': '2.062', 'learning_rate': '0.001', 'epoch': '0.4853', 'train/total_time_seconds': '365.3', 'train/time_per_step_avg': '0.4056', 'train/epoch_time_elapsed': '712.3', 'train/estimated_remaining_minutes': '0.6765'}
+  - 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, 65.03it/s]
+100%|██████████| 1000/1000 [13:21<00:00,  1.27i[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': '8.492', 'grad_norm': '1.922', 'learning_rate': '0.001', 'epoch': '0.4961', 'train/total_time_seconds': '373.4', 'train/time_per_step_avg': '0.4057', 'train/epoch_time_elapsed': '728.2', 'train/estimated_remaining_minutes': '0.5412'}
+{'loss': '8.474', 'grad_norm': '1.891', 'learning_rate': '0.001', 'epoch': '0.5069', 'train/total_time_seconds': '381.5', 'train/time_per_step_avg': '0.4059', 'train/epoch_time_elapsed': '743.9', 'train/estimated_remaining_minutes': '0.4059'}
+{'loss': '8.452', 'grad_norm': '1.844', 'learning_rate': '0.001', 'epoch': '0.5177', 'train/total_time_seconds': '389.7', 'train/time_per_step_avg': '0.406', 'train/epoch_time_elapsed': '759.8', 'train/estimated_remaining_minutes': '0.2706'}
+{'loss': '8.417', 'grad_norm': '1.891', 'learning_rate': '0.001', 'epoch': '0.5284', 'train/total_time_seconds': '397.8', 'train/time_per_step_avg': '0.4058', 'train/epoch_time_elapsed': '775.7', 'train/estimated_remaining_minutes': '0.1353'}
+{'loss': '8.381', 'grad_norm': '1.844', 'learning_rate': '0.001', 'epoch': '0.5392', 'train/total_time_seconds': '405.9', 'train/time_per_step_avg': '0.4063', 'train/epoch_time_elapsed': '791.6', 'train/estimated_remaining_minutes': '0'}
+  - 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': '2.094', 'eval_runtime': '9.59', 'eval_samples_per_second': '993.4', 'eval_steps_per_second': '7.821', 'epoch': '0.5392', 'train/total_time_seconds': '405.9', 'train/time_per_step_avg': '0.4063', 'train/epoch_time_elapsed': '801.2', 'train/estimated_remaining_minutes': '0'}
+  - 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, 76.03it/s]
+100%|██████████| 1000/1000 [13:21<00:00,  1.25it/s], ?it/s]
+{'train_runtime': '802.1', 'train_samples_per_second': '638.3', 'train_steps_per_second': '1.247', 'train_loss': '11.79', 'epoch': '0.5392', 'train/total_time_seconds': '405.9', 'train/time_per_step_avg': '0.4063', 'train/epoch_time_elapsed': '801.2', 'train/estimated_remaining_minutes': '0'}
+100%|██████████| 75/75 [00:09<00:00,  7.85it/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.
+  - 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, 79.14it/s]
+>>> FINISHED glu-waleed-21L successfully

zain/Activation/wandb/run-20260814_222456-wcvy5vyw/files/requirements.txt

@@ -0,0 +1,149 @@
+asttokens==3.0.1
+comm==0.2.3
+debugpy==1.8.21
+decorator==5.3.1
+executing==2.2.1
+nest-asyncio==1.6.0
+parso==0.8.7
+platformdirs==4.11.0
+psutil==7.2.2
+ptyprocess==0.7.0
+pure_eval==0.2.3
+Pygments==2.20.0
+pyzmq==27.1.0
+setuptools==83.0.0
+six==1.17.0
+tornado==6.5.7
+traitlets==5.15.0
+fsspec==2026.4.0
+wcwidth==0.8.2
+ipython_pygments_lexers==1.1.1
+jedi==0.20.0
+jupyter_core==5.9.1
+matplotlib-inline==0.2.2
+pexpect==4.9.0
+prompt_toolkit==3.0.53
+python-dateutil==2.9.0.post0
+stack_data==0.6.3
+wheel==0.47.0
+jupyter_client==8.9.1
+pip==26.1.2
+ipython==9.15.0
+ipykernel==7.2.0
+threadpoolctl==3.6.0
+pyparsing==3.3.2
+typing_extensions==4.15.0
+Jinja2==3.1.6
+narwhals==2.24.0
+kiwisolver==1.5.0
+joblib==1.5.3
+fonttools==4.63.0
+cycler==0.12.1
+scipy==1.17.1
+pandas==3.0.5
+contourpy==1.3.3
+scikit-learn==1.9.0
+matplotlib==3.11.1
+urllib3==2.7.0
+tqdm==4.70.0
+idna==3.18
+charset-normalizer==3.4.9
+certifi==2026.7.22
+requests==2.34.2
+seaborn==0.13.2
+uv==0.12.0
+shellingham==1.5.4
+mpmath==1.3.0
+attrs==26.1.0
+hf-xet==1.5.2
+nvidia-nccl-cu12==2.21.5
+MarkupSafe==3.0.3
+regex==2026.7.19
+importlib_metadata==9.0.0
+httpcore==1.0.9
+annotated-doc==0.0.5
+multidict==6.7.1
+aiohttp==3.14.3
+aiosignal==1.4.0
+xxhash==3.8.1
+aiohappyeyeballs==2.7.1
+mdurl==0.1.2
+cuda-toolkit==13.0.3.0
+networkx==3.6.1
+PyYAML==6.0.3
+nvidia-cufile==1.15.1.6
+typer==0.27.0
+torchaudio==2.6.0+cu124
+rich==15.0.0
+nvidia-cufft-cu12==11.2.1.3
+h11==0.16.0
+dill==0.4.1
+cuda-pathfinder==1.6.0
+filelock==3.29.0
+nvidia-nvtx-cu12==12.4.127
+httpx==0.28.1
+anyio==4.14.2
+numpy==2.4.4
+yarl==1.24.5
+click==8.4.2
+triton==3.2.0
+frozenlist==1.8.0
+zipp==4.1.0
+propcache==0.5.2
+tokenizers==0.22.2
+markdown-it-py==4.2.0
+nvidia-cuda-runtime==13.0.96
+cuda-bindings==13.3.1
+nvidia-cuda-cupti==13.0.85
+torch==2.6.0+cu124
+multiprocess==0.70.19
+pillow==12.2.0
+transformers==5.16.0.dev0
+wandb==0.28.1
+nvidia-curand==10.4.0.35
+sympy==1.13.1
+nvidia-cusparse==12.6.3.3
+nvidia-cuda-nvrtc==13.0.88
+typing-inspection==0.4.2
+nvidia-cusolver==12.0.4.66
+nvidia-cufft==12.0.0.61
+nvidia-cudnn-cu13==9.20.0.48
+nvidia-cublas==13.1.1.3
+pyarrow==25.0.0
+evaluate==0.4.6
+diffusers==0.39.0
+pydantic==2.13.4
+annotated-types==0.8.0
+protobuf==7.35.1
+sentry-sdk==2.66.1
+einops==0.8.2
+packaging==26.2
+nvidia-nvjitlink-cu12==12.4.127
+nvidia-curand-cu12==10.3.5.147
+nvidia-cusparselt-cu12==0.6.2
+nvidia-cusparse-cu12==12.3.1.170
+nvidia-cuda-runtime-cu12==12.4.127
+torchvision==0.21.0+cu124
+nvidia-cuda-nvrtc-cu12==12.4.127
+nvidia-cuda-cupti-cu12==12.4.127
+nvidia-cusolver-cu12==11.6.1.9
+nvidia-cublas-cu12==12.4.5.8
+nvidia-cudnn-cu12==9.1.0.70
+huggingface_hub==1.26.0
+datasets==5.0.1
+safetensors==0.8.0
+accelerate==1.14.0
+pydantic_core==2.46.4
+ninja==1.13.0
+autocommand==2.2.2
+backports.tarfile==1.2.0
+importlib_metadata==8.7.1
+jaraco.text==4.0.0
+jaraco.context==6.1.0
+jaraco.functools==4.4.0
+more-itertools==10.8.0
+packaging==26.0
+platformdirs==4.4.0
+tomli==2.4.0
+wheel==0.46.3
+zipp==3.23.0

zain/Activation/wandb/run-20260814_222456-wcvy5vyw/files/wandb-metadata.json

@@ -0,0 +1,101 @@
+{
+  "os":  "Linux-5.15.0-126-generic-x86_64-with-glibc2.35",
+  "python":  "CPython 3.11.15",
+  "startedAt":  "2026-08-14T22:24:56.823342Z",
+  "args":  [
+    "--config",
+    "/mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/configs/baseline.yaml",
+    "--variants",
+    "glu-waleed10",
+    "glu-silu-waleed10",
+    "mlp-waleed10",
+    "mlp-silu-waleed10",
+    "glu-waleed",
+    "glu-situglu_low",
+    "glu-waleedglu_low"
+  ],
+  "program":  "/mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/sweep.py",
+  "codePath":  "sweep.py",
+  "codePathLocal":  "sweep.py",
+  "git":  {
+    "remote":  "https://github.com/w-ahmad1a10/Activation.git",
+    "commit":  "3fb47c6943d6927eafc39bd399a3d3a520bae74a"
+  },
+  "email":  "deepnevro@gmail.com",
+  "root":  "/mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation",
+  "host":  "deeplens-k3s-node1",
+  "executable":  "/mnt/data/zainulabideen/zain-exp/notebooks/my_env/bin/python",
+  "cpu_count":  112,
+  "cpu_count_logical":  224,
+  "gpu":  "NVIDIA H100 80GB HBM3",
+  "gpu_count":  8,
+  "disk":  {
+    "/":  {
+      "total":  "1560765693952",
+      "used":  "709103636480"
+    }
+  },
+  "memory":  {
+    "total":  "2164089937920"
+  },
+  "gpu_nvidia":  [
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-39c684a5-fde6-83d7-1663-0859795881ae"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-68012e5a-38b6-b643-0ca6-62fb66720bf3"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-132944c4-b689-2b5f-89a4-d730401677ab"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-2df386cc-6d26-d0e2-7a2d-a057b0d95864"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-bfa16575-1d94-1aa2-4537-2c93433f42ef"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-bc6c3e3c-9b90-09ca-c034-774961847c54"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-00a441e1-7c95-e7d6-4c35-43d6b291aea9"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-1c4d29a2-4647-6fce-d8fc-0c5ecfbbd6ea"
+    }
+  ],
+  "cudaVersion":  "12.4",
+  "writerId":  "ry559cdhs8pu8maqeo36io5wxu2p8d4g"
+}

zain/Activation/wandb/run-20260814_222456-wcvy5vyw/files/wandb-summary.json

@@ -0,0 +1 @@
+{"train/loss":8.381436920166015,"_timestamp":1.7867471085870068e+09,"eval/steps_per_second":7.683,"_step":52,"train/grad_norm":1.84375,"train_samples_per_second":638.348,"_wandb":{"runtime":811},"train_runtime":802.0703,"train_loss":11.791388671875,"_runtime":811,"train/global_step":1000,"eval/loss":2.0943470001220703,"total_flos":5.420315836416e+15,"eval/samples_per_second":975.891,"train/epoch":0.5392289026691831,"eval/runtime":9.7624,"train_steps_per_second":1.247,"train/learning_rate":0.001}

zain/Activation/wandb/run-20260814_222456-wcvy5vyw/logs/debug-core.log

@@ -0,0 +1,62 @@
+{"time":"2026-08-14T21:31:53.539152008Z","level":"INFO","msg":"main: starting server","port-filename":"/tmp/tmp650y3sc2/port-1229423.txt","pid":1229423,"detached":false,"idle-timeout":600000000000,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false,"enable-dcgm-profiling":false}
+{"time":"2026-08-14T21:31:53.540676876Z","level":"INFO","msg":"server: will exit if parent process dies","ppid":1229423}
+{"time":"2026-08-14T21:31:53.540659035Z","level":"INFO","msg":"server: accepting connections","addr":{"Name":"/tmp/wandb-1229423-1230766-2604858008/socket","Net":"unix"}}
+{"time":"2026-08-14T21:31:53.717724383Z","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"1(@)"}
+{"time":"2026-08-14T21:31:53.802092564Z","level":"INFO","msg":"handleInformInit: received","streamId":"q9vcmje5","id":"1(@)"}
+{"time":"2026-08-14T21:31:54.062397411Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"q9vcmje5","id":"1(@)"}
+{"time":"2026-08-14T21:31:59.411924652Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"32s1adqk73rr"}
+{"time":"2026-08-14T21:45:48.396635188Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"32s1adqk73rr"}
+{"time":"2026-08-14T21:45:48.929834053Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"32s1adqk73rr"}
+{"time":"2026-08-14T21:45:48.931620539Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"q9vcmje5","id":"1(@)"}
+{"time":"2026-08-14T21:45:48.932491197Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"q9vcmje5","id":"1(@)"}
+{"time":"2026-08-14T21:45:49.464388506Z","level":"INFO","msg":"handleInformInit: received","streamId":"k24y90ws","id":"1(@)"}
+{"time":"2026-08-14T21:45:49.733853364Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"k24y90ws","id":"1(@)"}
+{"time":"2026-08-14T21:45:55.329792921Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"dr322rnt7xgx"}
+{"time":"2026-08-14T21:59:01.158537572Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"dr322rnt7xgx"}
+{"time":"2026-08-14T21:59:01.914569559Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"dr322rnt7xgx"}
+{"time":"2026-08-14T21:59:01.916156477Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"k24y90ws","id":"1(@)"}
+{"time":"2026-08-14T21:59:01.917534874Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"k24y90ws","id":"1(@)"}
+{"time":"2026-08-14T21:59:02.524868682Z","level":"INFO","msg":"handleInformInit: received","streamId":"npap5k3g","id":"1(@)"}
+{"time":"2026-08-14T21:59:02.786199875Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"npap5k3g","id":"1(@)"}
+{"time":"2026-08-14T21:59:08.192858634Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"2g99yy5pussw"}
+{"time":"2026-08-14T22:11:53.643379161Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"2g99yy5pussw"}
+{"time":"2026-08-14T22:11:54.173579992Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"2g99yy5pussw"}
+{"time":"2026-08-14T22:11:54.17490129Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"npap5k3g","id":"1(@)"}
+{"time":"2026-08-14T22:11:54.175479186Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"npap5k3g","id":"1(@)"}
+{"time":"2026-08-14T22:11:54.694802218Z","level":"INFO","msg":"handleInformInit: received","streamId":"a3607q94","id":"1(@)"}
+{"time":"2026-08-14T22:11:54.956642682Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"a3607q94","id":"1(@)"}
+{"time":"2026-08-14T22:12:00.441973218Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"raaqqqmajsfs"}
+{"time":"2026-08-14T22:24:55.662446346Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"raaqqqmajsfs"}
+{"time":"2026-08-14T22:24:56.296327526Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"raaqqqmajsfs"}
+{"time":"2026-08-14T22:24:56.297810539Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"a3607q94","id":"1(@)"}
+{"time":"2026-08-14T22:24:56.29856977Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"a3607q94","id":"1(@)"}
+{"time":"2026-08-14T22:24:56.82560196Z","level":"INFO","msg":"handleInformInit: received","streamId":"wcvy5vyw","id":"1(@)"}
+{"time":"2026-08-14T22:24:57.085772231Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"wcvy5vyw","id":"1(@)"}
+{"time":"2026-08-14T22:25:02.571913681Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"2nb8wusqu4fo"}
+{"time":"2026-08-14T22:38:28.612623637Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"2nb8wusqu4fo"}
+{"time":"2026-08-14T22:38:29.208727987Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"2nb8wusqu4fo"}
+{"time":"2026-08-14T22:38:29.210106262Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"wcvy5vyw","id":"1(@)"}
+{"time":"2026-08-14T22:38:29.210705504Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"wcvy5vyw","id":"1(@)"}
+{"time":"2026-08-14T22:38:29.839594174Z","level":"INFO","msg":"handleInformInit: received","streamId":"uuv4h936","id":"1(@)"}
+{"time":"2026-08-14T22:38:30.099444184Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"uuv4h936","id":"1(@)"}
+{"time":"2026-08-14T22:38:35.620666929Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"r0aouorh27ls"}
+{"time":"2026-08-14T22:52:17.704425638Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"r0aouorh27ls"}
+{"time":"2026-08-14T22:52:18.35448297Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"r0aouorh27ls"}
+{"time":"2026-08-14T22:52:18.355720686Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"uuv4h936","id":"1(@)"}
+{"time":"2026-08-14T22:52:18.356578583Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"uuv4h936","id":"1(@)"}
+{"time":"2026-08-14T22:52:18.899799881Z","level":"INFO","msg":"handleInformInit: received","streamId":"x87hwrtd","id":"1(@)"}
+{"time":"2026-08-14T22:52:19.159494071Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"x87hwrtd","id":"1(@)"}
+{"time":"2026-08-14T22:52:24.671083701Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"hn7lqjhi3ekb"}
+{"time":"2026-08-14T23:05:51.214192173Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"hn7lqjhi3ekb"}
+{"time":"2026-08-14T23:05:51.783107435Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"hn7lqjhi3ekb"}
+{"time":"2026-08-14T23:05:51.784410941Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"x87hwrtd","id":"1(@)"}
+{"time":"2026-08-14T23:05:51.78516667Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"x87hwrtd","id":"1(@)"}
+{"time":"2026-08-14T23:05:52.088192277Z","level":"INFO","msg":"handleInformTeardown: server teardown initiated","id":"1(@)"}
+{"time":"2026-08-14T23:05:52.088229648Z","level":"INFO","msg":"handleInformTeardown: server shutdown complete","id":"1(@)"}
+{"time":"2026-08-14T23:05:52.088239651Z","level":"INFO","msg":"connection: closing","id":"1(@)"}
+{"time":"2026-08-14T23:05:52.088243873Z","level":"INFO","msg":"server: is shutting down"}
+{"time":"2026-08-14T23:05:52.0882852Z","level":"INFO","msg":"connection: closed successfully","id":"1(@)"}
+{"time":"2026-08-14T23:05:52.088299095Z","level":"INFO","msg":"processOutgoingData: finished","id":"1(@)"}
+{"time":"2026-08-14T23:05:52.088345917Z","level":"INFO","msg":"connection: ManageConnectionData: connection closed","id":"1(@)"}
+{"time":"2026-08-14T23:05:52.088452051Z","level":"INFO","msg":"server: listener closed","addr":{"Name":"/tmp/wandb-1229423-1230766-2604858008/socket","Net":"unix"}}
+{"time":"2026-08-14T23:05:52.088520571Z","level":"INFO","msg":"server: all connections closed"}

zain/Activation/wandb/run-20260814_222456-wcvy5vyw/logs/debug-internal.log

@@ -0,0 +1,125 @@
+{"time":"2026-08-14T22:24:56.825723698Z","level":"INFO","msg":"wandb-core"}
+{"time":"2026-08-14T22:24:56.825860434Z","level":"INFO","msg":"stream: starting","core version":"0.28.1"}
+{"time":"2026-08-14T22:24:57.085596625Z","level":"INFO","msg":"stream: created new stream","id":"wcvy5vyw"}
+{"time":"2026-08-14T22:24:57.085675681Z","level":"INFO","msg":"handler: started"}
+{"time":"2026-08-14T22:24:57.085765252Z","level":"INFO","msg":"stream: started"}
+{"time":"2026-08-14T22:24:57.085777062Z","level":"INFO","msg":"writer: started","stream_id":"wcvy5vyw"}
+{"time":"2026-08-14T22:24:57.085800553Z","level":"INFO","msg":"sender: started"}
+{"time":"2026-08-14T22:24:57.57390664Z","level":"INFO","msg":"filestream: sending request","total_files":1,"console_offset":0,"console_lines":1}
+{"time":"2026-08-14T22:24:57.689803675Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:25:12.574037646Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":0,"events_lines":1,"console_offset":0,"console_lines":1,"uploaded_len":2}
+{"time":"2026-08-14T22:25:12.716448176Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:25:27.574260399Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":0,"history_lines":1,"events_offset":1,"events_lines":2,"console_offset":0,"console_lines":2}
+{"time":"2026-08-14T22:25:27.69115135Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:25:42.57432798Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":1,"history_lines":1,"events_offset":3,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-14T22:25:42.701645665Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:25:57.574852945Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":2,"history_lines":1,"events_offset":5,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-14T22:25:57.694991424Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:26:12.574234616Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":3,"history_lines":1,"events_offset":7,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-14T22:26:12.702125322Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:26:27.574556836Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":4,"history_lines":1,"events_offset":9,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-14T22:26:27.713695135Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:26:42.574592485Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":5,"history_lines":1,"events_offset":11,"events_lines":2,"console_offset":2,"console_lines":10}
+{"time":"2026-08-14T22:26:42.717599899Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:26:57.574987828Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":6,"history_lines":1,"events_offset":13,"events_lines":2,"console_offset":10,"console_lines":1}
+{"time":"2026-08-14T22:26:57.697000807Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:27:12.574348283Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":7,"history_lines":1,"events_offset":15,"events_lines":2,"console_offset":10,"console_lines":1}
+{"time":"2026-08-14T22:27:12.700155262Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:27:27.574749814Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":8,"history_lines":1,"events_offset":17,"events_lines":2,"console_offset":10,"console_lines":1}
+{"time":"2026-08-14T22:27:27.728833877Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:27:42.574886541Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":9,"history_lines":1,"events_offset":19,"events_lines":2,"console_offset":10,"console_lines":1}
+{"time":"2026-08-14T22:27:42.725241152Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:27:57.574361574Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":10,"history_lines":1,"events_offset":21,"events_lines":2,"console_offset":12,"console_lines":10}
+{"time":"2026-08-14T22:27:57.708700673Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:28:12.574682815Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":11,"history_lines":1,"events_offset":23,"events_lines":2,"console_offset":20,"console_lines":1}
+{"time":"2026-08-14T22:28:12.674665971Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:28:27.574667399Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":12,"history_lines":1,"events_offset":25,"events_lines":2,"console_offset":20,"console_lines":1}
+{"time":"2026-08-14T22:28:27.740964744Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:28:42.574290372Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":13,"history_lines":1,"events_offset":27,"events_lines":2,"console_offset":20,"console_lines":1}
+{"time":"2026-08-14T22:28:43.234719315Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:28:57.574853814Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":14,"history_lines":1,"events_offset":29,"events_lines":2,"console_offset":20,"console_lines":1}
+{"time":"2026-08-14T22:28:57.694726394Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:29:12.574668999Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":15,"history_lines":1,"events_offset":31,"events_lines":2,"console_offset":22,"console_lines":10}
+{"time":"2026-08-14T22:29:12.693938985Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:29:27.574781248Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":16,"history_lines":1,"events_offset":33,"events_lines":2,"console_offset":30,"console_lines":1}
+{"time":"2026-08-14T22:29:27.684922113Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:29:42.574792655Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":17,"history_lines":1,"events_offset":35,"events_lines":2,"console_offset":30,"console_lines":1}
+{"time":"2026-08-14T22:29:42.702364027Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:29:57.574828034Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":37,"events_lines":2,"console_offset":30,"console_lines":1}
+{"time":"2026-08-14T22:29:57.682234649Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:30:12.574145436Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":18,"history_lines":1,"events_offset":39,"events_lines":2,"console_offset":30,"console_lines":1}
+{"time":"2026-08-14T22:30:12.78125588Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:30:27.574409495Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":19,"history_lines":1,"events_offset":41,"events_lines":2,"console_offset":30,"console_lines":1}
+{"time":"2026-08-14T22:30:27.700286132Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:30:42.574579334Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":20,"history_lines":1,"events_offset":43,"events_lines":2,"console_offset":32,"console_lines":10}
+{"time":"2026-08-14T22:30:42.708397219Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:30:57.574936618Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":21,"history_lines":1,"events_offset":45,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-14T22:30:57.677385895Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:31:12.574052907Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":22,"history_lines":1,"events_offset":47,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-14T22:31:12.698626832Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:31:27.57497536Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":23,"history_lines":1,"events_offset":49,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-14T22:31:27.711296606Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:31:42.574875419Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":24,"history_lines":1,"events_offset":51,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-14T22:31:42.686140815Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:31:57.574985842Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":25,"history_lines":1,"events_offset":53,"events_lines":2,"console_offset":42,"console_lines":10}
+{"time":"2026-08-14T22:31:57.692228972Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:32:12.574537455Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":26,"history_lines":1,"events_offset":55,"events_lines":2,"console_offset":50,"console_lines":1}
+{"time":"2026-08-14T22:32:12.72236182Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:32:27.574897977Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":27,"history_lines":1,"events_offset":57,"events_lines":2,"console_offset":50,"console_lines":1}
+{"time":"2026-08-14T22:32:27.698118542Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:32:42.574300653Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":28,"history_lines":1,"events_offset":59,"events_lines":2,"console_offset":50,"console_lines":1}
+{"time":"2026-08-14T22:32:42.690560187Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:32:57.575083653Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":29,"history_lines":1,"events_offset":61,"events_lines":2,"console_offset":50,"console_lines":1}
+{"time":"2026-08-14T22:32:57.720449984Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:33:12.574474659Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":30,"history_lines":1,"events_offset":63,"events_lines":2,"console_offset":52,"console_lines":10}
+{"time":"2026-08-14T22:33:12.713825792Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:33:27.574130185Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":31,"history_lines":1,"events_offset":65,"events_lines":2,"console_offset":60,"console_lines":1}
+{"time":"2026-08-14T22:33:27.727843221Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:33:42.57400035Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":32,"history_lines":1,"events_offset":67,"events_lines":2,"console_offset":60,"console_lines":1}
+{"time":"2026-08-14T22:33:42.678772321Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:33:57.574860667Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":33,"history_lines":1,"events_offset":69,"events_lines":2,"console_offset":60,"console_lines":1}
+{"time":"2026-08-14T22:33:57.703764077Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:34:12.574978134Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":34,"history_lines":1,"events_offset":71,"events_lines":2,"console_offset":60,"console_lines":1}
+{"time":"2026-08-14T22:34:12.695794566Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:34:27.574270423Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":35,"history_lines":1,"events_offset":73,"events_lines":2,"console_offset":62,"console_lines":10}
+{"time":"2026-08-14T22:34:27.687851697Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:34:42.574471733Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":75,"events_lines":2,"console_offset":70,"console_lines":1}
+{"time":"2026-08-14T22:34:42.678295361Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:34:57.574760136Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":36,"history_lines":1,"events_offset":77,"events_lines":2,"console_offset":70,"console_lines":1}
+{"time":"2026-08-14T22:34:57.695664984Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:35:12.574728541Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":37,"history_lines":1,"events_offset":79,"events_lines":2,"console_offset":70,"console_lines":1}
+{"time":"2026-08-14T22:35:12.697377259Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:35:27.574941286Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":38,"history_lines":1,"events_offset":81,"events_lines":2,"console_offset":70,"console_lines":1}
+{"time":"2026-08-14T22:35:27.682936495Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:35:42.574971781Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":39,"history_lines":1,"events_offset":83,"events_lines":2,"console_offset":70,"console_lines":1}
+{"time":"2026-08-14T22:35:42.714357465Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:35:57.574263028Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":40,"history_lines":1,"events_offset":85,"events_lines":2,"console_offset":72,"console_lines":10}
+{"time":"2026-08-14T22:35:57.708124425Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:36:12.574958811Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":41,"history_lines":1,"events_offset":87,"events_lines":2,"console_offset":80,"console_lines":1}
+{"time":"2026-08-14T22:36:12.697898169Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:36:27.574130752Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":42,"history_lines":1,"events_offset":89,"events_lines":2,"console_offset":80,"console_lines":1}
+{"time":"2026-08-14T22:36:27.700264798Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:36:42.574300051Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":43,"history_lines":1,"events_offset":91,"events_lines":2,"console_offset":80,"console_lines":1}
+{"time":"2026-08-14T22:36:42.709508515Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:36:57.574652978Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":44,"history_lines":1,"events_offset":93,"events_lines":2,"console_offset":80,"console_lines":1}
+{"time":"2026-08-14T22:36:57.697730869Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:37:12.574917991Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":45,"history_lines":1,"events_offset":95,"events_lines":2,"console_offset":82,"console_lines":10}
+{"time":"2026-08-14T22:37:12.735095383Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:37:27.574633025Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":46,"history_lines":1,"events_offset":97,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-14T22:37:27.698944697Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:37:42.574415424Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":47,"history_lines":1,"events_offset":99,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-14T22:37:42.734335436Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:37:57.574721565Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":48,"history_lines":1,"events_offset":101,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-14T22:37:57.698777829Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:38:12.574423276Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":49,"history_lines":1,"events_offset":103,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-14T22:38:12.775133791Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:38:27.574276913Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":50,"history_lines":2,"events_offset":105,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-14T22:38:27.693401254Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:38:29.060944776Z","level":"INFO","msg":"fileTransfer: Close: file transfer manager closed"}
+{"time":"2026-08-14T22:38:29.061209871Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":52,"history_lines":1,"events_offset":107,"events_lines":1,"console_offset":92,"console_lines":18,"uploaded_len":3,"complete":true,"exit_code":0}
+{"time":"2026-08-14T22:38:29.206456668Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T22:38:29.207654339Z","level":"INFO","msg":"handler: operation stats","stats":{}}
+{"time":"2026-08-14T22:38:29.210126011Z","level":"INFO","msg":"stream: finishing up"}
+{"time":"2026-08-14T22:38:29.210150844Z","level":"INFO","msg":"handler: closed"}
+{"time":"2026-08-14T22:38:29.210267138Z","level":"INFO","msg":"sender: closed"}
+{"time":"2026-08-14T22:38:29.210270442Z","level":"INFO","msg":"stream: all finished"}

zain/Activation/wandb/run-20260814_222456-wcvy5vyw/logs/debug.log

@@ -0,0 +1,24 @@
+2026-08-14 22:24:56,824 INFO    MainThread:1229423 [wandb_init.py:setup_run_log_directory():729] Logging user logs to /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/wandb/run-20260814_222456-wcvy5vyw/logs/debug.log
+2026-08-14 22:24:56,824 INFO    MainThread:1229423 [wandb_init.py:setup_run_log_directory():730] Logging internal logs to /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/wandb/run-20260814_222456-wcvy5vyw/logs/debug-internal.log
+2026-08-14 22:24:56,824 INFO    MainThread:1229423 [wandb_init.py:init():772] calling init triggers
+2026-08-14 22:24:56,824 INFO    MainThread:1229423 [wandb_init.py:init():777] wandb.init called with sweep_config: {}
+config: {'_wandb': {}}
+2026-08-14 22:24:56,824 INFO    MainThread:1229423 [wandb_init.py:init():820] starting backend
+2026-08-14 22:24:56,824 INFO    MainThread:1229423 [wandb_init.py:init():835] sending inform_init request
+2026-08-14 22:24:57,086 INFO    MainThread:1229423 [wandb_init.py:init():840] backend started and connected
+2026-08-14 22:24:57,087 INFO    MainThread:1229423 [wandb_init.py:init():910] updated telemetry
+2026-08-14 22:24:57,094 INFO    MainThread:1229423 [wandb_init.py:init():933] communicating run to backend with 90.0 second timeout
+2026-08-14 22:24:57,491 INFO    MainThread:1229423 [wandb_init.py:init():978] starting run threads in backend
+2026-08-14 22:24:57,568 INFO    MainThread:1229423 [wandb_run.py:_console_start():2621] atexit reg
+2026-08-14 22:24:57,568 INFO    MainThread:1229423 [wandb_run.py:_redirect():2471] redirect: wrap_raw
+2026-08-14 22:24:57,568 INFO    MainThread:1229423 [wandb_run.py:_redirect():2540] Wrapping output streams.
+2026-08-14 22:24:57,568 INFO    MainThread:1229423 [wandb_run.py:_redirect():2563] Redirects installed.
+2026-08-14 22:24:57,569 INFO    MainThread:1229423 [wandb_init.py:init():1016] run started, returning control to user process
+2026-08-14 22:24:57,570 INFO    MainThread:1229423 [wandb_run.py:_config_callback():1346] config_cb None None {'transformers_version': '5.16.0.dev0', 'architectures': None, 'output_hidden_states': False, 'return_dict': True, 'dtype': None, 'chunk_size_feed_forward': 0, 'is_encoder_decoder': False, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'problem_type': None, 'vocab_size': 4096, 'hidden_size': 128, 'intermediate_size': 256, 'num_hidden_layers': 21, 'num_attention_heads': 4, 'num_key_value_heads': 4, 'hidden_act': 'silu', 'max_position_embeddings': 512, 'initializer_range': 0.02, 'rms_norm_eps': 1e-06, 'use_cache': False, 'pad_token_id': 0, 'bos_token_id': 1, 'eos_token_id': 2, 'pretraining_tp': 1, 'tie_word_embeddings': True, 'rope_parameters': {'rope_theta': 10000.0, 'rope_type': 'default'}, 'attention_bias': False, 'attention_dropout': 0.0, 'mlp_bias': False, 'head_dim': 32, '_name_or_path': '', 'tokenizer_name': 'w-ahmad/tiny-stories-tokenizer', 'mlp_type': 'glu', 'activation': 'waleed', 'waleed_beta': 10.0, 'model_type': 'tiny_llama', 'output_attentions': False, 'output_dir': 'out/glu-waleed-21L_run', 'per_device_train_batch_size': 128, 'num_train_epochs': 1, 'max_steps': 1000, 'learning_rate': 0.001, 'lr_scheduler_type': 'constant', 'lr_scheduler_kwargs': None, 'warmup_steps': 0, 'optim': 'adamw_torch', 'optim_args': None, 'weight_decay': 0.01, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'optim_target_modules': None, 'gradient_accumulation_steps': 4, 'average_tokens_across_devices': True, 'max_grad_norm': 1.0, 'label_smoothing_factor': 0.0, 'bf16': True, 'fp16': False, 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'gradient_checkpointing': False, 'gradient_checkpointing_kwargs': None, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'use_liger_kernel': False, 'liger_kernel_config': None, 'neftune_noise_alpha': None, 'torch_empty_cache_steps': None, 'auto_find_batch_size': False, 'logging_strategy': 'steps', 'logging_steps': 20, 'logging_first_step': False, 'log_on_each_node': True, 'logging_nan_inf_filter': True, 'include_num_input_tokens_seen': 'no', 'log_level': 'passive', 'log_level_replica': 'warning', 'disable_tqdm': False, 'report_to': ['wandb'], 'run_name': 'LM-glu-waleed-21L-4.0M-20260814-222456', 'project': 'huggingface', 'trackio_space_id': None, 'trackio_bucket_id': None, 'trackio_static_space_id': None, 'eval_strategy': 'steps', 'eval_steps': 60000, 'eval_delay': 0, 'per_device_eval_batch_size': 128, 'prediction_loss_only': False, 'eval_on_start': False, 'eval_do_concat_batches': True, 'eval_use_gather_object': False, 'eval_accumulation_steps': None, 'include_for_metrics': [], 'batch_eval_metrics': False, 'save_only_model': False, 'save_strategy': 'steps', 'save_steps': 100, 'save_on_each_node': False, 'save_total_limit': None, 'enable_jit_checkpoint': False, 'push_to_hub': False, 'hub_token': '<HUB_TOKEN>', 'hub_private_repo': None, 'hub_model_id': 'w-ahmad/A-glu-waleed-21L', 'hub_strategy': 'every_save', 'hub_always_push': False, 'hub_revision': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'restore_callback_states_from_checkpoint': False, 'full_determinism': False, 'seed': 42, 'data_seed': 42, 'use_cpu': False, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}, 'parallelism_config': None, 'dataloader_drop_last': False, 'dataloader_num_workers': 0, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'dataloader_prefetch_factor': None, 'dataloader_multiprocessing_context': None, 'dataloader_in_order': True, 'remove_unused_columns': False, 'label_names': None, 'train_sampling_strategy': 'random', 'length_column_name': 'length', 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'ddp_static_graph': None, 'ddp_backend': None, 'ddp_timeout': 1800, 'fsdp': None, 'fsdp_config': None, 'deepspeed': None, 'debug': [], 'skip_memory_metrics': True, 'do_train': False, 'do_eval': True, 'do_predict': False, 'resume_from_checkpoint': None, 'local_rank': -1}
+2026-08-14 22:24:57,572 INFO    MainThread:1229423 [wandb_config.py:__setitem__():155] [no run ID] config set model/num_parameters = 3970432 - <bound method Run._config_callback of <wandb.sdk.wandb_run.Run object at 0x15298883e390>>
+2026-08-14 22:24:57,573 INFO    MainThread:1229423 [wandb_run.py:_config_callback():1346] config_cb model/num_parameters 3970432 None
+2026-08-14 22:38:28,611 INFO    MainThread:1229423 [wandb_run.py:_finish():2383] finishing run deepnevro-deepnevro/research1/wcvy5vyw
+2026-08-14 22:38:28,612 INFO    MainThread:1229423 [wandb_run.py:_atexit_cleanup():2588] got exitcode: 0
+2026-08-14 22:38:28,612 INFO    MainThread:1229423 [wandb_run.py:_restore():2570] restore
+2026-08-14 22:38:28,612 INFO    MainThread:1229423 [wandb_run.py:_restore():2576] restore done
+2026-08-14 22:38:29,209 INFO    MainThread:1229423 [wandb_run.py:_footer_sync_info():3993] logging synced files