Auto upload zain 2026-08-14T20:43:33.111422 (part 4)

.gitattributes

@@ -48,3 +48,5 @@ zain/Activation/wandb/run-20260814_200503-bdhnr27j/run-bdhnr27j.wandb filter=lfs
 zain/Activation/wandb/run-20260814_201615-w8mr4w6k/run-w8mr4w6k.wandb filter=lfs diff=lfs merge=lfs -text
 zain/Activation/wandb/run-20260814_202128-twgo1e89/run-twgo1e89.wandb filter=lfs diff=lfs merge=lfs -text
 zain/Activation/wandb/run-20260814_202806-g2ch84q3/run-g2ch84q3.wandb filter=lfs diff=lfs merge=lfs -text
+zain/Activation/wandb/run-20260814_203803-7p5140bq/run-7p5140bq.wandb filter=lfs diff=lfs merge=lfs -text
+zain/Activation/wandb/run-20260814_203945-j86czemo/run-j86czemo.wandb filter=lfs diff=lfs merge=lfs -text

zain/Activation/out/mlp-waleed10-9L_run/config.json

@@ -0,0 +1,36 @@
+{
+  "activation": "waleed10",
+  "architectures": [
+    "TinyLlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "dtype": "bfloat16",
+  "eos_token_id": 2,
+  "head_dim": 32,
+  "hidden_act": "silu",
+  "hidden_size": 128,
+  "initializer_range": 0.02,
+  "intermediate_size": 256,
+  "max_position_embeddings": 512,
+  "mlp_bias": false,
+  "mlp_type": "mlp",
+  "model_type": "tiny_llama",
+  "num_attention_heads": 4,
+  "num_hidden_layers": 9,
+  "num_key_value_heads": 4,
+  "pad_token_id": 0,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-06,
+  "rope_parameters": {
+    "rope_theta": 10000.0,
+    "rope_type": "default"
+  },
+  "tie_word_embeddings": true,
+  "tokenizer_name": "w-ahmad/tiny-stories-tokenizer",
+  "transformers_version": "5.16.0.dev0",
+  "use_cache": false,
+  "vocab_size": 4096,
+  "waleed_beta": 10.0
+}

zain/Activation/out/mlp-waleed10-9L_run/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8f2b230a08f5bc03bac3d5730be00f65d1714089fc0d0e7cc13b806564e116d8
+size 4010544

zain/Activation/out/mlp-waleed10-9L_run/tokenizer_config.json

@@ -0,0 +1,13 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": "<|endoftext|>",
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "is_local": false,
+  "local_files_only": false,
+  "model_max_length": 1024,
+  "pad_token": "<|endoftext|>",
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": "<|endoftext|>"
+}

zain/Activation/out/mlp-waleed10-9L_run/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e619779e59dc19818520b0351cd8871bdc29ee6df9cd5004d2bf8cc1836e250e
+size 4920

zain/Activation/out/mlp-waleed10-9L_run/training_log.jsonl

@@ -44,3 +44,10 @@
 {"step": 880, "epoch": 0.4743935309973046, "timestamp": 1786739883.5160599, "loss": 2.68295841217041, "grad_norm": 1.140625, "learning_rate": 0.001, "train/total_time_seconds": 263.4235079474747, "train/time_per_step_avg": 0.2746262482553721, "train/epoch_time_elapsed": 596.1756766997278, "train/estimated_remaining_minutes": 0.5986897907897153}
 {"step": 900, "epoch": 0.48517520215633425, "timestamp": 1786739897.1374428, "loss": 2.6740900039672852, "grad_norm": 0.82421875, "learning_rate": 0.001, "train/total_time_seconds": 269.418474663049, "train/time_per_step_avg": 0.28068523917347193, "train/epoch_time_elapsed": 609.7970554046333, "train/estimated_remaining_minutes": 0.4989231012278685}
 {"step": 920, "epoch": 0.49595687331536387, "timestamp": 1786739910.96125, "loss": 2.6633071899414062, "grad_norm": 0.71875, "learning_rate": 0.001, "train/total_time_seconds": 275.5517046675086, "train/time_per_step_avg": 0.28259021110832694, "train/epoch_time_elapsed": 623.6208665408194, "train/estimated_remaining_minutes": 0.39935029661957766}
+{"step": 940, "epoch": 0.5067385444743935, "timestamp": 1786739924.6428165, "loss": 2.6555158615112306, "grad_norm": 0.63671875, "learning_rate": 0.001, "train/total_time_seconds": 281.58670626580715, "train/time_per_step_avg": 0.29087681483477357, "train/epoch_time_elapsed": 637.302432730794, "train/estimated_remaining_minutes": 0.29956032581468844}
+{"step": 960, "epoch": 0.5175202156334232, "timestamp": 1786739938.3910303, "loss": 2.6521677017211913, "grad_norm": 0.98046875, "learning_rate": 0.001, "train/total_time_seconds": 287.67976446822286, "train/time_per_step_avg": 0.2981896402314305, "train/epoch_time_elapsed": 651.0506465211511, "train/estimated_remaining_minutes": 0.19977761421404364}
+{"step": 980, "epoch": 0.5283018867924528, "timestamp": 1786739952.1133928, "loss": 2.6380289077758787, "grad_norm": 0.859375, "learning_rate": 0.001, "train/total_time_seconds": 293.7678536660969, "train/time_per_step_avg": 0.30344345718622207, "train/epoch_time_elapsed": 664.7730091884732, "train/estimated_remaining_minutes": 0.09992103866193773}
+{"step": 1000, "epoch": 0.5390835579514824, "timestamp": 1786739965.7814448, "loss": 2.6369375228881835, "grad_norm": 1.0390625, "learning_rate": 0.001, "train/total_time_seconds": 299.73355290293694, "train/time_per_step_avg": 0.30315078239887955, "train/epoch_time_elapsed": 678.4410613812506, "train/estimated_remaining_minutes": 0.0}
+{"step": 1000, "epoch": 0.5390835579514824, "timestamp": 1786739974.9520266, "eval_loss": 2.6366770267486572, "eval_runtime": 9.1687, "eval_samples_per_second": 1039.073, "eval_steps_per_second": 2.072, "train/total_time_seconds": 299.73355290293694, "train/time_per_step_avg": 0.30315078239887955, "train/epoch_time_elapsed": 687.6116429083049, "train/estimated_remaining_minutes": 0.0}
+{"step": 1000, "epoch": 0.5390835579514824, "timestamp": 1786739975.0022418, "train_runtime": 688.4499, "train_samples_per_second": 743.7, "train_steps_per_second": 1.453, "total_flos": 2323107545088000.0, "train_loss": 3.3107347106933593, "train/total_time_seconds": 299.73355290293694, "train/time_per_step_avg": 0.30315078239887955, "train/epoch_time_elapsed": 687.6618567891419, "train/estimated_remaining_minutes": 0.0}
+{"step": 1000, "epoch": 0.5390835579514824, "timestamp": 1786739983.9396794, "eval_loss": 2.6366770267486572, "eval_runtime": 8.9348, "eval_samples_per_second": 1066.275, "eval_steps_per_second": 2.127, "train/total_time_seconds": 299.73355290293694, "train/time_per_step_avg": 0.30315078239887955, "train/epoch_time_elapsed": 696.5992951057851, "train/estimated_remaining_minutes": 0.0}

zain/Activation/wandb/debug-internal.log

@@ -1,11 +1,39 @@
-{"time":"2026-08-14T20:38:03.474745278Z","level":"INFO","msg":"wandb-core"}
-{"time":"2026-08-14T20:38:03.474866577Z","level":"INFO","msg":"stream: starting","core version":"0.28.1"}
-{"time":"2026-08-14T20:38:03.739205178Z","level":"INFO","msg":"stream: created new stream","id":"7p5140bq"}
-{"time":"2026-08-14T20:38:03.73930164Z","level":"INFO","msg":"handler: started"}
-{"time":"2026-08-14T20:38:03.739394556Z","level":"INFO","msg":"stream: started"}
-{"time":"2026-08-14T20:38:03.739418374Z","level":"INFO","msg":"writer: started","stream_id":"7p5140bq"}
-{"time":"2026-08-14T20:38:03.739441138Z","level":"INFO","msg":"sender: started"}
-{"time":"2026-08-14T20:38:04.1733199Z","level":"INFO","msg":"filestream: sending request","total_files":1,"console_offset":0,"console_lines":1}
-{"time":"2026-08-14T20:38:04.349662463Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-14T20:38:19.173810778Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":0,"history_lines":1,"events_offset":0,"events_lines":1,"console_offset":0,"console_lines":2,"uploaded_len":2}
-{"time":"2026-08-14T20:38:19.494112066Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:39:45.020404854Z","level":"INFO","msg":"wandb-core"}
+{"time":"2026-08-14T20:39:45.020535959Z","level":"INFO","msg":"stream: starting","core version":"0.28.1"}
+{"time":"2026-08-14T20:39:45.279239855Z","level":"INFO","msg":"stream: created new stream","id":"j86czemo"}
+{"time":"2026-08-14T20:39:45.279331465Z","level":"INFO","msg":"handler: started"}
+{"time":"2026-08-14T20:39:45.27940313Z","level":"INFO","msg":"stream: started"}
+{"time":"2026-08-14T20:39:45.279417021Z","level":"INFO","msg":"writer: started","stream_id":"j86czemo"}
+{"time":"2026-08-14T20:39:45.279439783Z","level":"INFO","msg":"sender: started"}
+{"time":"2026-08-14T20:39:45.743788117Z","level":"INFO","msg":"filestream: sending request","total_files":1,"console_offset":0,"console_lines":1}
+{"time":"2026-08-14T20:39:45.85449847Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:40:00.74416275Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":0,"history_lines":1,"events_offset":0,"events_lines":1,"console_offset":0,"console_lines":2,"uploaded_len":2}
+{"time":"2026-08-14T20:40:00.977965075Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:40:15.744587204Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":1,"history_lines":1,"events_offset":1,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-14T20:40:15.928506721Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:40:30.743869324Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":2,"history_lines":1,"events_offset":3,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-14T20:40:30.917560257Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:40:45.744229552Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":3,"history_lines":1,"events_offset":5,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-14T20:40:45.875347202Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:41:00.744908115Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":4,"history_lines":1,"events_offset":7,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-14T20:41:00.918429764Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:41:15.744313716Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":5,"history_lines":1,"events_offset":9,"events_lines":2,"console_offset":2,"console_lines":10}
+{"time":"2026-08-14T20:41:15.966956122Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:41:30.744909744Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":6,"history_lines":1,"events_offset":11,"events_lines":2,"console_offset":10,"console_lines":1}
+{"time":"2026-08-14T20:41:30.903737277Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:41:45.744811784Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":7,"history_lines":1,"events_offset":13,"events_lines":2,"console_offset":10,"console_lines":1}
+{"time":"2026-08-14T20:41:45.881876523Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:42:00.744201973Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":8,"history_lines":1,"events_offset":15,"events_lines":2,"console_offset":10,"console_lines":1}
+{"time":"2026-08-14T20:42:00.896435129Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:42:15.743927337Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":9,"history_lines":1,"events_offset":17,"events_lines":2,"console_offset":10,"console_lines":1}
+{"time":"2026-08-14T20:42:15.934416786Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:42:30.744737861Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":10,"history_lines":1,"events_offset":19,"events_lines":2,"console_offset":12,"console_lines":10}
+{"time":"2026-08-14T20:42:30.892376698Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:42:45.744822111Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":11,"history_lines":2,"events_offset":21,"events_lines":2,"console_offset":20,"console_lines":1}
+{"time":"2026-08-14T20:42:45.971531834Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:43:00.744614977Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":13,"history_lines":1,"events_offset":23,"events_lines":2,"console_offset":20,"console_lines":1}
+{"time":"2026-08-14T20:43:00.949080571Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:43:15.745651796Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":14,"history_lines":1,"events_offset":25,"events_lines":2,"console_offset":20,"console_lines":1}
+{"time":"2026-08-14T20:43:15.911365748Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:43:30.744257431Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":15,"history_lines":1,"events_offset":27,"events_lines":2,"console_offset":22,"console_lines":10}
+{"time":"2026-08-14T20:43:30.956144577Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}

zain/Activation/wandb/debug.log

@@ -1,20 +1,19 @@
-2026-08-14 20:38:03,473 INFO    MainThread:171901 [wandb_init.py:setup_run_log_directory():729] Logging user logs to /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/wandb/run-20260814_203803-7p5140bq/logs/debug.log
-2026-08-14 20:38:03,473 INFO    MainThread:171901 [wandb_init.py:setup_run_log_directory():730] Logging internal logs to /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/wandb/run-20260814_203803-7p5140bq/logs/debug-internal.log
-2026-08-14 20:38:03,473 INFO    MainThread:171901 [wandb_init.py:init():772] calling init triggers
-2026-08-14 20:38:03,473 INFO    MainThread:171901 [wandb_init.py:init():777] wandb.init called with sweep_config: {}
+2026-08-14 20:39:45,019 INFO    MainThread:409188 [wandb_init.py:setup_run_log_directory():729] Logging user logs to /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/wandb/run-20260814_203945-j86czemo/logs/debug.log
+2026-08-14 20:39:45,019 INFO    MainThread:409188 [wandb_init.py:setup_run_log_directory():730] Logging internal logs to /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/wandb/run-20260814_203945-j86czemo/logs/debug-internal.log
+2026-08-14 20:39:45,019 INFO    MainThread:409188 [wandb_init.py:init():772] calling init triggers
+2026-08-14 20:39:45,019 INFO    MainThread:409188 [wandb_init.py:init():777] wandb.init called with sweep_config: {}
 config: {'_wandb': {}}
-2026-08-14 20:38:03,473 INFO    MainThread:171901 [wandb_init.py:init():820] starting backend
-2026-08-14 20:38:03,473 INFO    MainThread:171901 [wandb_init.py:init():826] Connected to an existing wandb-core service via WANDB_SERVICE
-2026-08-14 20:38:03,473 INFO    MainThread:171901 [wandb_init.py:init():835] sending inform_init request
-2026-08-14 20:38:03,739 INFO    MainThread:171901 [wandb_init.py:init():840] backend started and connected
-2026-08-14 20:38:03,741 INFO    MainThread:171901 [wandb_init.py:init():910] updated telemetry
-2026-08-14 20:38:03,748 INFO    MainThread:171901 [wandb_init.py:init():933] communicating run to backend with 90.0 second timeout
-2026-08-14 20:38:04,083 INFO    MainThread:171901 [wandb_init.py:init():978] starting run threads in backend
-2026-08-14 20:38:04,164 INFO    MainThread:171901 [wandb_run.py:_console_start():2621] atexit reg
-2026-08-14 20:38:04,164 INFO    MainThread:171901 [wandb_run.py:_redirect():2471] redirect: wrap_raw
-2026-08-14 20:38:04,164 INFO    MainThread:171901 [wandb_run.py:_redirect():2540] Wrapping output streams.
-2026-08-14 20:38:04,164 INFO    MainThread:171901 [wandb_run.py:_redirect():2563] Redirects installed.
-2026-08-14 20:38:04,165 INFO    MainThread:171901 [wandb_init.py:init():1016] run started, returning control to user process
-2026-08-14 20:38:04,166 INFO    MainThread:171901 [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_run', 'per_device_train_batch_size': 64, 'num_train_epochs': 1, 'max_steps': 2500, 'learning_rate': 0.0007, 'lr_scheduler_type': 'constant_with_warmup', 'lr_scheduler_kwargs': None, 'warmup_steps': 500, '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-20260814-203802', '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': 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/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 20:38:04,172 INFO    MainThread:171901 [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 0x1505b01c2e10>>
-2026-08-14 20:38:04,172 INFO    MainThread:171901 [wandb_run.py:_config_callback():1346] config_cb model/num_parameters 16934016 None
+2026-08-14 20:39:45,019 INFO    MainThread:409188 [wandb_init.py:init():820] starting backend
+2026-08-14 20:39:45,019 INFO    MainThread:409188 [wandb_init.py:init():835] sending inform_init request
+2026-08-14 20:39:45,279 INFO    MainThread:409188 [wandb_init.py:init():840] backend started and connected
+2026-08-14 20:39:45,280 INFO    MainThread:409188 [wandb_init.py:init():910] updated telemetry
+2026-08-14 20:39:45,287 INFO    MainThread:409188 [wandb_init.py:init():933] communicating run to backend with 90.0 second timeout
+2026-08-14 20:39:45,662 INFO    MainThread:409188 [wandb_init.py:init():978] starting run threads in backend
+2026-08-14 20:39:45,738 INFO    MainThread:409188 [wandb_run.py:_console_start():2621] atexit reg
+2026-08-14 20:39:45,738 INFO    MainThread:409188 [wandb_run.py:_redirect():2471] redirect: wrap_raw
+2026-08-14 20:39:45,738 INFO    MainThread:409188 [wandb_run.py:_redirect():2540] Wrapping output streams.
+2026-08-14 20:39:45,738 INFO    MainThread:409188 [wandb_run.py:_redirect():2563] Redirects installed.
+2026-08-14 20:39:45,739 INFO    MainThread:409188 [wandb_init.py:init():1016] run started, returning control to user process
+2026-08-14 20:39:45,740 INFO    MainThread:409188 [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': 9, '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-waleed10-9L_run', 'per_device_train_batch_size': 512, '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': 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-mlp-silu-waleed10-9L-2.0M-20260814-203944', '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': 512, '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-waleed10-9L', '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 20:39:45,742 INFO    MainThread:409188 [wandb_config.py:__setitem__():155] [no run ID] config set model/num_parameters = 2001280 - <bound method Run._config_callback of <wandb.sdk.wandb_run.Run object at 0x153f4a14ab50>>
+2026-08-14 20:39:45,742 INFO    MainThread:409188 [wandb_run.py:_config_callback():1346] config_cb model/num_parameters 2001280 None

zain/Activation/wandb/run-20260814_200425-npumf6lz/logs/debug-core.log

@@ -19,3 +19,10 @@
 {"time":"2026-08-14T20:28:06.637357786Z","level":"INFO","msg":"handleInformInit: received","streamId":"g2ch84q3","id":"1(@)"}
 {"time":"2026-08-14T20:28:06.894723818Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"g2ch84q3","id":"1(@)"}
 {"time":"2026-08-14T20:28:12.337494383Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"3viso7y2z8wp"}
+{"time":"2026-08-14T20:39:43.955226503Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"3viso7y2z8wp"}
+{"time":"2026-08-14T20:39:44.505666869Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"3viso7y2z8wp"}
+{"time":"2026-08-14T20:39:44.507136651Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"g2ch84q3","id":"1(@)"}
+{"time":"2026-08-14T20:39:44.507897996Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"g2ch84q3","id":"1(@)"}
+{"time":"2026-08-14T20:39:45.020279672Z","level":"INFO","msg":"handleInformInit: received","streamId":"j86czemo","id":"1(@)"}
+{"time":"2026-08-14T20:39:45.27941156Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"j86czemo","id":"1(@)"}
+{"time":"2026-08-14T20:39:50.74200388Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"ibw5o8e0zw49"}

zain/Activation/wandb/run-20260814_201615-w8mr4w6k/logs/debug-core.log

@@ -19,3 +19,10 @@
 {"time":"2026-08-14T20:28:06.637357786Z","level":"INFO","msg":"handleInformInit: received","streamId":"g2ch84q3","id":"1(@)"}
 {"time":"2026-08-14T20:28:06.894723818Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"g2ch84q3","id":"1(@)"}
 {"time":"2026-08-14T20:28:12.337494383Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"3viso7y2z8wp"}
+{"time":"2026-08-14T20:39:43.955226503Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"3viso7y2z8wp"}
+{"time":"2026-08-14T20:39:44.505666869Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"3viso7y2z8wp"}
+{"time":"2026-08-14T20:39:44.507136651Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"g2ch84q3","id":"1(@)"}
+{"time":"2026-08-14T20:39:44.507897996Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"g2ch84q3","id":"1(@)"}
+{"time":"2026-08-14T20:39:45.020279672Z","level":"INFO","msg":"handleInformInit: received","streamId":"j86czemo","id":"1(@)"}
+{"time":"2026-08-14T20:39:45.27941156Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"j86czemo","id":"1(@)"}
+{"time":"2026-08-14T20:39:50.74200388Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"ibw5o8e0zw49"}

zain/Activation/wandb/run-20260814_202806-g2ch84q3/files/config.yaml

@@ -0,0 +1,439 @@
+_name_or_path:
+    value: ""
+_wandb:
+    value:
+        cli_version: 0.28.1
+        e:
+            xs4pfgh48fmwfr7akyadjs1xiecxnuk9:
+                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: "696174620672"
+                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: "2026-08-14T20:28:06.635059Z"
+                writerId: xs4pfgh48fmwfr7akyadjs1xiecxnuk9
+        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: 1
+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-waleed10-9L
+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: 2001280
+model_type:
+    value: tiny_llama
+neftune_noise_alpha:
+    value: null
+num_attention_heads:
+    value: 4
+num_hidden_layers:
+    value: 9
+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-waleed10-9L_run
+output_hidden_states:
+    value: false
+pad_token_id:
+    value: 0
+parallelism_config:
+    value: null
+per_device_eval_batch_size:
+    value: 512
+per_device_train_batch_size:
+    value: 512
+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-waleed10-9L-2.0M-20260814-202806
+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_202806-g2ch84q3/files/output.log

@@ -88,5 +88,23 @@ Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 140
   - 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, 167.04it/s]
- 92%|█████████▏| 922/1000 [10:24<00:52,  1.49it/s]?, ?it/s]
+100%|██████████| 1000/1000 [11:27<00:00,  1.46i[transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
 {'loss': '2.663', 'grad_norm': '0.7188', 'learning_rate': '0.001', 'epoch': '0.496', 'train/total_time_seconds': '275.6', 'train/time_per_step_avg': '0.2826', 'train/epoch_time_elapsed': '623.6', 'train/estimated_remaining_minutes': '0.3994'}
+{'loss': '2.656', 'grad_norm': '0.6367', 'learning_rate': '0.001', 'epoch': '0.5067', 'train/total_time_seconds': '281.6', 'train/time_per_step_avg': '0.2909', 'train/epoch_time_elapsed': '637.3', 'train/estimated_remaining_minutes': '0.2996'}
+{'loss': '2.652', 'grad_norm': '0.9805', 'learning_rate': '0.001', 'epoch': '0.5175', 'train/total_time_seconds': '287.7', 'train/time_per_step_avg': '0.2982', 'train/epoch_time_elapsed': '651.1', 'train/estimated_remaining_minutes': '0.1998'}
+{'loss': '2.638', 'grad_norm': '0.8594', 'learning_rate': '0.001', 'epoch': '0.5283', 'train/total_time_seconds': '293.8', 'train/time_per_step_avg': '0.3034', 'train/epoch_time_elapsed': '664.8', 'train/estimated_remaining_minutes': '0.09992'}
+{'loss': '2.637', 'grad_norm': '1.039', 'learning_rate': '0.001', 'epoch': '0.5391', 'train/total_time_seconds': '299.7', 'train/time_per_step_avg': '0.3032', 'train/epoch_time_elapsed': '678.4', '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': '2.637', 'eval_runtime': '9.169', 'eval_samples_per_second': '1039', 'eval_steps_per_second': '2.072', 'epoch': '0.5391', 'train/total_time_seconds': '299.7', 'train/time_per_step_avg': '0.3032', 'train/epoch_time_elapsed': '687.6', '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, 107.44it/s]
+100%|██████████| 1000/1000 [11:27<00:00,  1.45it/s], ?it/s]
+{'train_runtime': '688.4', 'train_samples_per_second': '743.7', 'train_steps_per_second': '1.453', 'train_loss': '3.311', 'epoch': '0.5391', 'train/total_time_seconds': '299.7', 'train/time_per_step_avg': '0.3032', 'train/epoch_time_elapsed': '687.7', 'train/estimated_remaining_minutes': '0'}
+100%|██████████| 19/19 [00:08<00:00,  2.33it/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.
+  - 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, 165.85it/s]
+>>> FINISHED mlp-waleed10-9L successfully

zain/Activation/wandb/run-20260814_202806-g2ch84q3/files/wandb-summary.json

@@ -0,0 +1 @@
+{"train_loss":3.3107347106933593,"train/learning_rate":0.001,"train_samples_per_second":743.7,"_wandb":{"runtime":696},"train/epoch":0.5390835579514824,"_timestamp":1.7867399839392445e+09,"_runtime":696,"eval/loss":2.6366770267486572,"train/global_step":1000,"total_flos":2.323107545088e+15,"_step":52,"eval/runtime":8.9348,"train_steps_per_second":1.453,"train/loss":2.6369375228881835,"train/grad_norm":1.0390625,"eval/samples_per_second":1066.275,"eval/steps_per_second":2.127,"train_runtime":688.4499}

zain/Activation/wandb/run-20260814_202806-g2ch84q3/logs/debug-core.log

@@ -19,3 +19,10 @@
 {"time":"2026-08-14T20:28:06.637357786Z","level":"INFO","msg":"handleInformInit: received","streamId":"g2ch84q3","id":"1(@)"}
 {"time":"2026-08-14T20:28:06.894723818Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"g2ch84q3","id":"1(@)"}
 {"time":"2026-08-14T20:28:12.337494383Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"3viso7y2z8wp"}
+{"time":"2026-08-14T20:39:43.955226503Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"3viso7y2z8wp"}
+{"time":"2026-08-14T20:39:44.505666869Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"3viso7y2z8wp"}
+{"time":"2026-08-14T20:39:44.507136651Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"g2ch84q3","id":"1(@)"}
+{"time":"2026-08-14T20:39:44.507897996Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"g2ch84q3","id":"1(@)"}
+{"time":"2026-08-14T20:39:45.020279672Z","level":"INFO","msg":"handleInformInit: received","streamId":"j86czemo","id":"1(@)"}
+{"time":"2026-08-14T20:39:45.27941156Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"j86czemo","id":"1(@)"}
+{"time":"2026-08-14T20:39:50.74200388Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"ibw5o8e0zw49"}

zain/Activation/wandb/run-20260814_202806-g2ch84q3/logs/debug-internal.log

@@ -89,3 +89,21 @@
 {"time":"2026-08-14T20:38:07.581146769Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
 {"time":"2026-08-14T20:38:22.341908397Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":44,"history_lines":1,"events_offset":79,"events_lines":2,"console_offset":80,"console_lines":1}
 {"time":"2026-08-14T20:38:22.507136925Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:38:37.34151095Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":45,"history_lines":1,"events_offset":81,"events_lines":2,"console_offset":83,"console_lines":9}
+{"time":"2026-08-14T20:38:37.525773463Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:38:52.341625562Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":46,"history_lines":1,"events_offset":83,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-14T20:38:52.4884038Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:39:07.341707427Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":47,"history_lines":1,"events_offset":85,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-14T20:39:07.541501066Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:39:22.341230177Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":48,"history_lines":1,"events_offset":87,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-14T20:39:22.487837117Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:39:37.341843026Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":49,"history_lines":3,"events_offset":89,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-14T20:39:37.555554822Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:39:44.372813974Z","level":"INFO","msg":"fileTransfer: Close: file transfer manager closed"}
+{"time":"2026-08-14T20:39:44.3730353Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":52,"history_lines":1,"events_offset":91,"events_lines":1,"console_offset":92,"console_lines":18,"uploaded_len":3,"complete":true,"exit_code":0}
+{"time":"2026-08-14T20:39:44.503492138Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:39:44.504681145Z","level":"INFO","msg":"handler: operation stats","stats":{}}
+{"time":"2026-08-14T20:39:44.507161532Z","level":"INFO","msg":"stream: finishing up"}
+{"time":"2026-08-14T20:39:44.507176089Z","level":"INFO","msg":"handler: closed"}
+{"time":"2026-08-14T20:39:44.507242589Z","level":"INFO","msg":"sender: closed"}
+{"time":"2026-08-14T20:39:44.507246036Z","level":"INFO","msg":"stream: all finished"}

zain/Activation/wandb/run-20260814_202806-g2ch84q3/logs/debug.log

@@ -17,3 +17,8 @@ config: {'_wandb': {}}
 2026-08-14 20:28:07,337 INFO    MainThread:409188 [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': 9, '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-waleed10-9L_run', 'per_device_train_batch_size': 512, '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': 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-mlp-waleed10-9L-2.0M-20260814-202806', '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': 512, '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-waleed10-9L', '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 20:28:07,339 INFO    MainThread:409188 [wandb_config.py:__setitem__():155] [no run ID] config set model/num_parameters = 2001280 - <bound method Run._config_callback of <wandb.sdk.wandb_run.Run object at 0x153f142af950>>
 2026-08-14 20:28:07,339 INFO    MainThread:409188 [wandb_run.py:_config_callback():1346] config_cb model/num_parameters 2001280 None
+2026-08-14 20:39:43,954 INFO    MainThread:409188 [wandb_run.py:_finish():2383] finishing run deepnevro-deepnevro/research1/g2ch84q3
+2026-08-14 20:39:43,954 INFO    MainThread:409188 [wandb_run.py:_atexit_cleanup():2588] got exitcode: 0
+2026-08-14 20:39:43,955 INFO    MainThread:409188 [wandb_run.py:_restore():2570] restore
+2026-08-14 20:39:43,955 INFO    MainThread:409188 [wandb_run.py:_restore():2576] restore done
+2026-08-14 20:39:44,506 INFO    MainThread:409188 [wandb_run.py:_footer_sync_info():3993] logging synced files

zain/Activation/wandb/run-20260814_202806-g2ch84q3/run-g2ch84q3.wandb

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f02950dda40849af1d6ca8e6cb85e83f5f4719d7f9c0d45024165da08867952a
-size 425984
+oid sha256:88495cf23fa022409d83d0b5707d890709378cdd766bc20872752e5c98fefbfd
+size 511047

zain/Activation/wandb/run-20260814_203803-7p5140bq/files/output.log

@@ -1,4 +1,85 @@
-  3%|▎         | 66/2500 [00:24<14:16,  2.84it/s]
+  4%|▍         | 100/2500 [00:37<14:49,  2.70it/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': '8.3', 'grad_norm': '2.078', 'learning_rate': '2.66e-05', 'epoch': '0.001348', 'train/total_time_seconds': '6.624', 'train/time_per_step_avg': '0.3312', 'train/epoch_time_elapsed': '7.719', 'train/estimated_remaining_minutes': '13.69'}
 {'loss': '8.048', 'grad_norm': '1.352', 'learning_rate': '5.46e-05', 'epoch': '0.002696', 'train/total_time_seconds': '13.15', 'train/time_per_step_avg': '0.3287', 'train/epoch_time_elapsed': '15.27', 'train/estimated_remaining_minutes': '13.48'}
 {'loss': '7.838', 'grad_norm': '1.273', 'learning_rate': '8.26e-05', 'epoch': '0.004044', 'train/total_time_seconds': '19.68', 'train/time_per_step_avg': '0.328', 'train/epoch_time_elapsed': '22.8', 'train/estimated_remaining_minutes': '13.34'}
+{'loss': '7.573', 'grad_norm': '1.266', 'learning_rate': '0.0001106', 'epoch': '0.005393', 'train/total_time_seconds': '26.13', 'train/time_per_step_avg': '0.3267', 'train/epoch_time_elapsed': '30.24', 'train/estimated_remaining_minutes': '13.18'}
+{'loss': '7.233', 'grad_norm': '1.195', 'learning_rate': '0.0001386', 'epoch': '0.006741', 'train/total_time_seconds': '32.49', 'train/time_per_step_avg': '0.3249', 'train/epoch_time_elapsed': '37.59', 'train/estimated_remaining_minutes': '13'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 16.09it/s]
+  8%|▊         | 200/2500 [01:14<13:57,  2.75it/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': '6.861', 'grad_norm': '1.133', 'learning_rate': '0.0001666', 'epoch': '0.008089', 'train/total_time_seconds': '39.01', 'train/time_per_step_avg': '0.3239', 'train/epoch_time_elapsed': '45.41', 'train/estimated_remaining_minutes': '12.9'}
+{'loss': '6.518', 'grad_norm': '0.9492', 'learning_rate': '0.0001946', 'epoch': '0.009437', 'train/total_time_seconds': '45.4', 'train/time_per_step_avg': '0.3225', 'train/epoch_time_elapsed': '52.8', 'train/estimated_remaining_minutes': '12.75'}
+{'loss': '6.24', 'grad_norm': '0.8008', 'learning_rate': '0.0002226', 'epoch': '0.01079', 'train/total_time_seconds': '51.77', 'train/time_per_step_avg': '0.3209', 'train/epoch_time_elapsed': '60.16', 'train/estimated_remaining_minutes': '12.62'}
+{'loss': '6.035', 'grad_norm': '0.7461', 'learning_rate': '0.0002506', 'epoch': '0.01213', 'train/total_time_seconds': '58.12', 'train/time_per_step_avg': '0.3199', 'train/epoch_time_elapsed': '67.5', 'train/estimated_remaining_minutes': '12.48'}
+{'loss': '5.874', 'grad_norm': '0.5547', 'learning_rate': '0.0002786', 'epoch': '0.01348', 'train/total_time_seconds': '64.6', 'train/time_per_step_avg': '0.3211', 'train/epoch_time_elapsed': '74.97', 'train/estimated_remaining_minutes': '12.38'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 16.10it/s]
+ 12%|█▏        | 300/2500 [01:49<14:50,  2.47it/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': '5.691', 'grad_norm': '0.498', 'learning_rate': '0.0003066', 'epoch': '0.01483', 'train/total_time_seconds': '70.89', 'train/time_per_step_avg': '0.3187', 'train/epoch_time_elapsed': '82.55', 'train/estimated_remaining_minutes': '12.24'}
+{'loss': '5.474', 'grad_norm': '1.219', 'learning_rate': '0.0003346', 'epoch': '0.01618', 'train/total_time_seconds': '76.44', 'train/time_per_step_avg': '0.3104', 'train/epoch_time_elapsed': '89.09', 'train/estimated_remaining_minutes': '12'}
+{'loss': '5.259', 'grad_norm': '1.508', 'learning_rate': '0.0003626', 'epoch': '0.01753', 'train/total_time_seconds': '82.15', 'train/time_per_step_avg': '0.3038', 'train/epoch_time_elapsed': '95.8', 'train/estimated_remaining_minutes': '11.8'}
+{'loss': '5.03', 'grad_norm': '0.7969', 'learning_rate': '0.0003906', 'epoch': '0.01887', 'train/total_time_seconds': '87.44', 'train/time_per_step_avg': '0.2932', 'train/epoch_time_elapsed': '102.1', 'train/estimated_remaining_minutes': '11.55'}
+{'loss': '4.798', 'grad_norm': '0.8438', 'learning_rate': '0.0004186', 'epoch': '0.02022', 'train/total_time_seconds': '93.92', 'train/time_per_step_avg': '0.2932', 'train/epoch_time_elapsed': '109.6', 'train/estimated_remaining_minutes': '11.48'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 15.55it/s]
+ 16%|█▌        | 400/2500 [02:27<13:29,  2.59it/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': '4.604', 'grad_norm': '1.508', 'learning_rate': '0.0004466', 'epoch': '0.02157', 'train/total_time_seconds': '100.3', 'train/time_per_step_avg': '0.2944', 'train/epoch_time_elapsed': '117.3', 'train/estimated_remaining_minutes': '11.39'}
+{'loss': '4.491', 'grad_norm': '0.9648', 'learning_rate': '0.0004746', 'epoch': '0.02292', 'train/total_time_seconds': '106.8', 'train/time_per_step_avg': '0.3038', 'train/epoch_time_elapsed': '124.8', 'train/estimated_remaining_minutes': '11.31'}
+{'loss': '4.327', 'grad_norm': '0.6641', 'learning_rate': '0.0005026', 'epoch': '0.02427', 'train/total_time_seconds': '113.3', 'train/time_per_step_avg': '0.3118', 'train/epoch_time_elapsed': '132.3', 'train/estimated_remaining_minutes': '11.23'}
+{'loss': '4.182', 'grad_norm': '0.8633', 'learning_rate': '0.0005306', 'epoch': '0.02562', 'train/total_time_seconds': '119.9', 'train/time_per_step_avg': '0.325', 'train/epoch_time_elapsed': '139.9', 'train/estimated_remaining_minutes': '11.15'}
+{'loss': '4.036', 'grad_norm': '0.8242', 'learning_rate': '0.0005586', 'epoch': '0.02696', 'train/total_time_seconds': '126.4', 'train/time_per_step_avg': '0.3252', 'train/epoch_time_elapsed': '147.4', 'train/estimated_remaining_minutes': '11.06'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 15.16it/s]
+ 20%|██        | 500/2500 [03:05<12:14,  2.72it/s][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '3.919', 'grad_norm': '0.7148', 'learning_rate': '0.0005866', 'epoch': '0.02831', 'train/total_time_seconds': '132.9', 'train/time_per_step_avg': '0.326', 'train/epoch_time_elapsed': '155.2', 'train/estimated_remaining_minutes': '10.97'}
+{'loss': '3.856', 'grad_norm': '0.7656', 'learning_rate': '0.0006146', 'epoch': '0.02966', 'train/total_time_seconds': '139.4', 'train/time_per_step_avg': '0.3261', 'train/epoch_time_elapsed': '162.7', 'train/estimated_remaining_minutes': '10.88'}
+{'loss': '3.75', 'grad_norm': '0.7734', 'learning_rate': '0.0006426', 'epoch': '0.03101', 'train/total_time_seconds': '145.9', 'train/time_per_step_avg': '0.3253', 'train/epoch_time_elapsed': '170.2', 'train/estimated_remaining_minutes': '10.78'}
+{'loss': '3.669', 'grad_norm': '0.6484', 'learning_rate': '0.0006706', 'epoch': '0.03236', 'train/total_time_seconds': '152.5', 'train/time_per_step_avg': '0.3252', 'train/epoch_time_elapsed': '177.8', 'train/estimated_remaining_minutes': '10.69'}
+{'loss': '3.598', 'grad_norm': '0.8828', 'learning_rate': '0.0006986', 'epoch': '0.0337', 'train/total_time_seconds': '159', 'train/time_per_step_avg': '0.3253', 'train/epoch_time_elapsed': '185.3', 'train/estimated_remaining_minutes': '10.6'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 15.78it/s]
+ 24%|██▍       | 600/2500 [03:43<12:41,  2.50it/s][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '3.529', 'grad_norm': '0.5625', 'learning_rate': '0.0007', 'epoch': '0.03505', 'train/total_time_seconds': '165.5', 'train/time_per_step_avg': '0.3253', 'train/epoch_time_elapsed': '193.1', 'train/estimated_remaining_minutes': '10.5'}
+{'loss': '3.449', 'grad_norm': '0.5664', 'learning_rate': '0.0007', 'epoch': '0.0364', 'train/total_time_seconds': '171.9', 'train/time_per_step_avg': '0.325', 'train/epoch_time_elapsed': '200.6', 'train/estimated_remaining_minutes': '10.4'}
+{'loss': '3.398', 'grad_norm': '0.4531', 'learning_rate': '0.0007', 'epoch': '0.03775', 'train/total_time_seconds': '178.4', 'train/time_per_step_avg': '0.3257', 'train/epoch_time_elapsed': '208.1', 'train/estimated_remaining_minutes': '10.3'}
+{'loss': '3.337', 'grad_norm': '0.5234', 'learning_rate': '0.0007', 'epoch': '0.0391', 'train/total_time_seconds': '185.3', 'train/time_per_step_avg': '0.3288', 'train/epoch_time_elapsed': '216', 'train/estimated_remaining_minutes': '10.23'}
+{'loss': '3.304', 'grad_norm': '0.4883', 'learning_rate': '0.0007', 'epoch': '0.04044', 'train/total_time_seconds': '191.8', 'train/time_per_step_avg': '0.3286', 'train/epoch_time_elapsed': '223.5', 'train/estimated_remaining_minutes': '10.12'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 16.88it/s]
+ 28%|██▊       | 700/2500 [04:21<10:49,  2.77it/s][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '3.237', 'grad_norm': '0.5508', 'learning_rate': '0.0007', 'epoch': '0.04179', 'train/total_time_seconds': '198.2', 'train/time_per_step_avg': '0.328', 'train/epoch_time_elapsed': '231.2', 'train/estimated_remaining_minutes': '10.02'}
+{'loss': '3.211', 'grad_norm': '0.4707', 'learning_rate': '0.0007', 'epoch': '0.04314', 'train/total_time_seconds': '204.7', 'train/time_per_step_avg': '0.3279', 'train/epoch_time_elapsed': '238.6', 'train/estimated_remaining_minutes': '9.916'}
+{'loss': '3.154', 'grad_norm': '0.4688', 'learning_rate': '0.0007', 'epoch': '0.04449', 'train/total_time_seconds': '211.3', 'train/time_per_step_avg': '0.3283', 'train/epoch_time_elapsed': '246.2', 'train/estimated_remaining_minutes': '9.816'}
+{'loss': '3.126', 'grad_norm': '0.4121', 'learning_rate': '0.0007', 'epoch': '0.04584', 'train/total_time_seconds': '217.8', 'train/time_per_step_avg': '0.3242', 'train/epoch_time_elapsed': '253.7', 'train/estimated_remaining_minutes': '9.714'}
+{'loss': '3.08', 'grad_norm': '0.4492', 'learning_rate': '0.0007', 'epoch': '0.04719', 'train/total_time_seconds': '224.3', 'train/time_per_step_avg': '0.3248', 'train/epoch_time_elapsed': '261.3', 'train/estimated_remaining_minutes': '9.614'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 11.98it/s]
+ 32%|███▏      | 800/2500 [04:59<10:26,  2.71it/s][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '3.041', 'grad_norm': '0.3906', 'learning_rate': '0.0007', 'epoch': '0.04853', 'train/total_time_seconds': '230.9', 'train/time_per_step_avg': '0.3264', 'train/epoch_time_elapsed': '269.2', 'train/estimated_remaining_minutes': '9.514'}
+{'loss': '3.003', 'grad_norm': '0.459', 'learning_rate': '0.0007', 'epoch': '0.04988', 'train/total_time_seconds': '237.4', 'train/time_per_step_avg': '0.3268', 'train/epoch_time_elapsed': '276.7', 'train/estimated_remaining_minutes': '9.411'}
+{'loss': '2.985', 'grad_norm': '0.4062', 'learning_rate': '0.0007', 'epoch': '0.05123', 'train/total_time_seconds': '243.9', 'train/time_per_step_avg': '0.3259', 'train/epoch_time_elapsed': '284.2', 'train/estimated_remaining_minutes': '9.305'}
+{'loss': '2.986', 'grad_norm': '0.4883', 'learning_rate': '0.0007', 'epoch': '0.05258', 'train/total_time_seconds': '250.4', 'train/time_per_step_avg': '0.3263', 'train/epoch_time_elapsed': '291.7', 'train/estimated_remaining_minutes': '9.202'}
+{'loss': '2.932', 'grad_norm': '0.4746', 'learning_rate': '0.0007', 'epoch': '0.05393', 'train/total_time_seconds': '256.9', 'train/time_per_step_avg': '0.3259', 'train/epoch_time_elapsed': '299.3', 'train/estimated_remaining_minutes': '9.099'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 15.60it/s]
+ 36%|███▌      | 894/2500 [05:34<10:00,  2.67it/s]?, ?it/s]
+{'loss': '2.916', 'grad_norm': '0.416', 'learning_rate': '0.0007', 'epoch': '0.05527', 'train/total_time_seconds': '263.5', 'train/time_per_step_avg': '0.3258', 'train/epoch_time_elapsed': '307.1', 'train/estimated_remaining_minutes': '8.997'}
+{'loss': '2.886', 'grad_norm': '0.373', 'learning_rate': '0.0007', 'epoch': '0.05662', 'train/total_time_seconds': '269.9', 'train/time_per_step_avg': '0.325', 'train/epoch_time_elapsed': '314.5', 'train/estimated_remaining_minutes': '8.89'}
+{'loss': '2.84', 'grad_norm': '0.4395', 'learning_rate': '0.0007', 'epoch': '0.05797', 'train/total_time_seconds': '276.3', 'train/time_per_step_avg': '0.3241', 'train/epoch_time_elapsed': '321.9', 'train/estimated_remaining_minutes': '8.78'}
+{'loss': '2.818', 'grad_norm': '0.4395', 'learning_rate': '0.0007', 'epoch': '0.05932', 'train/total_time_seconds': '283', 'train/time_per_step_avg': '0.3265', 'train/epoch_time_elapsed': '329.8', 'train/estimated_remaining_minutes': '8.684'}

zain/Activation/wandb/run-20260814_203803-7p5140bq/logs/debug-internal.log

@@ -9,3 +9,45 @@
 {"time":"2026-08-14T20:38:04.349662463Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
 {"time":"2026-08-14T20:38:19.173810778Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":0,"history_lines":1,"events_offset":0,"events_lines":1,"console_offset":0,"console_lines":2,"uploaded_len":2}
 {"time":"2026-08-14T20:38:19.494112066Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:38:34.174170719Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":1,"history_lines":2,"events_offset":1,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-14T20:38:34.355801404Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:38:49.173731969Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":3,"history_lines":2,"events_offset":3,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-14T20:38:49.307056378Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:39:04.174302993Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":5,"history_lines":2,"events_offset":5,"events_lines":2,"console_offset":2,"console_lines":11}
+{"time":"2026-08-14T20:39:04.384804376Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:39:19.173425986Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":7,"history_lines":3,"events_offset":7,"events_lines":2,"console_offset":10,"console_lines":1}
+{"time":"2026-08-14T20:39:19.315426459Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:39:34.173836176Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":10,"history_lines":2,"events_offset":9,"events_lines":2,"console_offset":13,"console_lines":10}
+{"time":"2026-08-14T20:39:34.336054489Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:39:49.173909734Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":12,"history_lines":2,"events_offset":11,"events_lines":2,"console_offset":20,"console_lines":1}
+{"time":"2026-08-14T20:39:49.369403865Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:40:04.174090891Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":14,"history_lines":2,"events_offset":13,"events_lines":2,"console_offset":20,"console_lines":1}
+{"time":"2026-08-14T20:40:04.568325682Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:40:19.173701072Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":16,"history_lines":2,"events_offset":15,"events_lines":2,"console_offset":23,"console_lines":11}
+{"time":"2026-08-14T20:40:19.332900945Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:40:34.17386923Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":18,"history_lines":2,"events_offset":17,"events_lines":2,"console_offset":30,"console_lines":1}
+{"time":"2026-08-14T20:40:34.316709884Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:40:49.174178438Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":20,"history_lines":2,"events_offset":19,"events_lines":2,"console_offset":34,"console_lines":9}
+{"time":"2026-08-14T20:40:49.339535632Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:41:04.174385674Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":22,"history_lines":2,"events_offset":21,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-14T20:41:04.400662878Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:41:19.17386313Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":24,"history_lines":2,"events_offset":23,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-14T20:41:19.286145363Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:41:34.17364169Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":26,"history_lines":2,"events_offset":25,"events_lines":2,"console_offset":43,"console_lines":11}
+{"time":"2026-08-14T20:41:34.344272617Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:41:49.17398552Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":28,"history_lines":2,"events_offset":27,"events_lines":2,"console_offset":50,"console_lines":1}
+{"time":"2026-08-14T20:41:49.282803705Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:42:04.174101565Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":30,"history_lines":2,"events_offset":29,"events_lines":2,"console_offset":54,"console_lines":9}
+{"time":"2026-08-14T20:42:04.346991656Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:42:19.174468763Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":32,"history_lines":2,"events_offset":31,"events_lines":2,"console_offset":60,"console_lines":1}
+{"time":"2026-08-14T20:42:19.311742849Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:42:34.173676954Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":34,"history_lines":2,"events_offset":33,"events_lines":2,"console_offset":60,"console_lines":1}
+{"time":"2026-08-14T20:42:34.291767713Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:42:49.17381101Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":36,"history_lines":2,"events_offset":35,"events_lines":2,"console_offset":63,"console_lines":11}
+{"time":"2026-08-14T20:42:49.328336145Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:43:04.173698143Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":38,"history_lines":2,"events_offset":37,"events_lines":2,"console_offset":70,"console_lines":1}
+{"time":"2026-08-14T20:43:04.319999371Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:43:19.173436098Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":40,"history_lines":2,"events_offset":39,"events_lines":2,"console_offset":74,"console_lines":9}
+{"time":"2026-08-14T20:43:19.295112626Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:43:34.174211982Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":42,"history_lines":2,"events_offset":41,"events_lines":2,"console_offset":80,"console_lines":1}
+{"time":"2026-08-14T20:43:34.312402397Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}

zain/Activation/wandb/run-20260814_203945-j86czemo/files/output.log

@@ -0,0 +1,33 @@
+ 10%|█         | 100/1000 [01:09<10:31,  1.43it/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': '7.157', 'grad_norm': '1.008', 'learning_rate': '0.001', 'epoch': '0.01078', 'train/total_time_seconds': '6.079', 'train/time_per_step_avg': '0.304', 'train/epoch_time_elapsed': '14.06', 'train/estimated_remaining_minutes': '4.965'}
+{'loss': '5.922', 'grad_norm': '0.6406', 'learning_rate': '0.001', 'epoch': '0.02156', 'train/total_time_seconds': '12.34', 'train/time_per_step_avg': '0.3084', 'train/epoch_time_elapsed': '27.79', 'train/estimated_remaining_minutes': '4.935'}
+{'loss': '5.383', 'grad_norm': '0.5781', 'learning_rate': '0.001', 'epoch': '0.03235', 'train/total_time_seconds': '18.61', 'train/time_per_step_avg': '0.3102', 'train/epoch_time_elapsed': '41.65', 'train/estimated_remaining_minutes': '4.859'}
+{'loss': '4.915', 'grad_norm': '0.6562', 'learning_rate': '0.001', 'epoch': '0.04313', 'train/total_time_seconds': '24.7', 'train/time_per_step_avg': '0.3088', 'train/epoch_time_elapsed': '55.18', 'train/estimated_remaining_minutes': '4.735'}
+{'loss': '4.547', 'grad_norm': '1.57', 'learning_rate': '0.001', 'epoch': '0.05391', 'train/total_time_seconds': '31.02', 'train/time_per_step_avg': '0.3102', 'train/epoch_time_elapsed': '69.03', 'train/estimated_remaining_minutes': '4.653'}
+  - 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, 168.45it/s]
+ 20%|██        | 200/1000 [02:17<09:06,  1.46it/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': '4.307', 'grad_norm': '0.7188', 'learning_rate': '0.001', 'epoch': '0.06469', 'train/total_time_seconds': '37.31', 'train/time_per_step_avg': '0.3123', 'train/epoch_time_elapsed': '82.89', 'train/estimated_remaining_minutes': '4.561'}
+{'loss': '4.099', 'grad_norm': '1.266', 'learning_rate': '0.001', 'epoch': '0.07547', 'train/total_time_seconds': '43.42', 'train/time_per_step_avg': '0.3108', 'train/epoch_time_elapsed': '96.4', 'train/estimated_remaining_minutes': '4.446'}
+{'loss': '3.924', 'grad_norm': '0.6914', 'learning_rate': '0.001', 'epoch': '0.08625', 'train/total_time_seconds': '49.61', 'train/time_per_step_avg': '0.31', 'train/epoch_time_elapsed': '110.2', 'train/estimated_remaining_minutes': '4.341'}
+{'loss': '3.787', 'grad_norm': '0.5234', 'learning_rate': '0.001', 'epoch': '0.09704', 'train/total_time_seconds': '55.81', 'train/time_per_step_avg': '0.311', 'train/epoch_time_elapsed': '123.8', 'train/estimated_remaining_minutes': '4.237'}
+{'loss': '3.674', 'grad_norm': '0.5195', 'learning_rate': '0.001', 'epoch': '0.1078', 'train/total_time_seconds': '62', 'train/time_per_step_avg': '0.3098', 'train/epoch_time_elapsed': '137.5', 'train/estimated_remaining_minutes': '4.134'}
+  - 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, 168.70it/s]
+ 30%|███       | 300/1000 [03:26<07:56,  1.47it/s][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '3.575', 'grad_norm': '0.7188', 'learning_rate': '0.001', 'epoch': '0.1186', 'train/total_time_seconds': '68.26', 'train/time_per_step_avg': '0.3095', 'train/epoch_time_elapsed': '151.2', 'train/estimated_remaining_minutes': '4.034'}
+{'loss': '3.497', 'grad_norm': '0.4707', 'learning_rate': '0.001', 'epoch': '0.1294', 'train/total_time_seconds': '74.39', 'train/time_per_step_avg': '0.3097', 'train/epoch_time_elapsed': '165', 'train/estimated_remaining_minutes': '3.926'}
+{'loss': '3.417', 'grad_norm': '0.5156', 'learning_rate': '0.001', 'epoch': '0.1402', 'train/total_time_seconds': '80.68', 'train/time_per_step_avg': '0.3107', 'train/epoch_time_elapsed': '178.8', 'train/estimated_remaining_minutes': '3.827'}
+{'loss': '3.354', 'grad_norm': '1.203', 'learning_rate': '0.001', 'epoch': '0.1509', 'train/total_time_seconds': '86.9', 'train/time_per_step_avg': '0.3109', 'train/epoch_time_elapsed': '192.4', 'train/estimated_remaining_minutes': '3.724'}
+{'loss': '3.294', 'grad_norm': '0.5938', 'learning_rate': '0.001', 'epoch': '0.1617', 'train/total_time_seconds': '93.09', 'train/time_per_step_avg': '0.3109', 'train/epoch_time_elapsed': '206.1', 'train/estimated_remaining_minutes': '3.62'}
+  - 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, 173.81it/s]
+ 34%|███▍      | 342/1000 [03:54<07:34,  1.45it/s]?, ?it/s]
+{'loss': '3.241', 'grad_norm': '1.055', 'learning_rate': '0.001', 'epoch': '0.1725', 'train/total_time_seconds': '99.27', 'train/time_per_step_avg': '0.3101', 'train/epoch_time_elapsed': '219.9', 'train/estimated_remaining_minutes': '3.516'}
+{'loss': '3.187', 'grad_norm': '0.6523', 'learning_rate': '0.001', 'epoch': '0.1833', 'train/total_time_seconds': '105.4', 'train/time_per_step_avg': '0.3096', 'train/epoch_time_elapsed': '233.5', 'train/estimated_remaining_minutes': '3.409'}

zain/Activation/wandb/run-20260814_203945-j86czemo/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_203945-j86czemo/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":  "2026-08-14T20:39:45.018091Z",
+  "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":  "696197320704"
+    }
+  },
+  "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":  "d0qv8t55nt5dj13xc4z90g53tv25vz77"
+}

zain/Activation/wandb/run-20260814_203945-j86czemo/logs/debug-core.log

@@ -0,0 +1,28 @@
+{"time":"2026-08-14T20:04:25.06988208Z","level":"INFO","msg":"main: starting server","port-filename":"/tmp/tmp4_27mh86/port-409188.txt","pid":409188,"detached":false,"idle-timeout":600000000000,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false,"enable-dcgm-profiling":false}
+{"time":"2026-08-14T20:04:25.070357566Z","level":"INFO","msg":"server: will exit if parent process dies","ppid":409188}
+{"time":"2026-08-14T20:04:25.070338944Z","level":"INFO","msg":"server: accepting connections","addr":{"Name":"/tmp/wandb-409188-410559-1836801570/socket","Net":"unix"}}
+{"time":"2026-08-14T20:04:25.24795932Z","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"1(@)"}
+{"time":"2026-08-14T20:04:25.504127391Z","level":"INFO","msg":"handleInformInit: received","streamId":"npumf6lz","id":"1(@)"}
+{"time":"2026-08-14T20:04:25.765319769Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"npumf6lz","id":"1(@)"}
+{"time":"2026-08-14T20:04:31.109816098Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"d0p7d03ygyti"}
+{"time":"2026-08-14T20:16:14.526898083Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"d0p7d03ygyti"}
+{"time":"2026-08-14T20:16:15.115854524Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"d0p7d03ygyti"}
+{"time":"2026-08-14T20:16:15.117591712Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"npumf6lz","id":"1(@)"}
+{"time":"2026-08-14T20:16:15.118263697Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"npumf6lz","id":"1(@)"}
+{"time":"2026-08-14T20:16:15.759561877Z","level":"INFO","msg":"handleInformInit: received","streamId":"w8mr4w6k","id":"1(@)"}
+{"time":"2026-08-14T20:16:16.020082435Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"w8mr4w6k","id":"1(@)"}
+{"time":"2026-08-14T20:16:21.531557612Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"ewwa8fnip5gl"}
+{"time":"2026-08-14T20:28:05.375497777Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"ewwa8fnip5gl"}
+{"time":"2026-08-14T20:28:06.097599346Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"ewwa8fnip5gl"}
+{"time":"2026-08-14T20:28:06.099103114Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"w8mr4w6k","id":"1(@)"}
+{"time":"2026-08-14T20:28:06.099811223Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"w8mr4w6k","id":"1(@)"}
+{"time":"2026-08-14T20:28:06.637357786Z","level":"INFO","msg":"handleInformInit: received","streamId":"g2ch84q3","id":"1(@)"}
+{"time":"2026-08-14T20:28:06.894723818Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"g2ch84q3","id":"1(@)"}
+{"time":"2026-08-14T20:28:12.337494383Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"3viso7y2z8wp"}
+{"time":"2026-08-14T20:39:43.955226503Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"3viso7y2z8wp"}
+{"time":"2026-08-14T20:39:44.505666869Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"3viso7y2z8wp"}
+{"time":"2026-08-14T20:39:44.507136651Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"g2ch84q3","id":"1(@)"}
+{"time":"2026-08-14T20:39:44.507897996Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"g2ch84q3","id":"1(@)"}
+{"time":"2026-08-14T20:39:45.020279672Z","level":"INFO","msg":"handleInformInit: received","streamId":"j86czemo","id":"1(@)"}
+{"time":"2026-08-14T20:39:45.27941156Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"j86czemo","id":"1(@)"}
+{"time":"2026-08-14T20:39:50.74200388Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"ibw5o8e0zw49"}

zain/Activation/wandb/run-20260814_203945-j86czemo/logs/debug-internal.log

@@ -0,0 +1,39 @@
+{"time":"2026-08-14T20:39:45.020404854Z","level":"INFO","msg":"wandb-core"}
+{"time":"2026-08-14T20:39:45.020535959Z","level":"INFO","msg":"stream: starting","core version":"0.28.1"}
+{"time":"2026-08-14T20:39:45.279239855Z","level":"INFO","msg":"stream: created new stream","id":"j86czemo"}
+{"time":"2026-08-14T20:39:45.279331465Z","level":"INFO","msg":"handler: started"}
+{"time":"2026-08-14T20:39:45.27940313Z","level":"INFO","msg":"stream: started"}
+{"time":"2026-08-14T20:39:45.279417021Z","level":"INFO","msg":"writer: started","stream_id":"j86czemo"}
+{"time":"2026-08-14T20:39:45.279439783Z","level":"INFO","msg":"sender: started"}
+{"time":"2026-08-14T20:39:45.743788117Z","level":"INFO","msg":"filestream: sending request","total_files":1,"console_offset":0,"console_lines":1}
+{"time":"2026-08-14T20:39:45.85449847Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:40:00.74416275Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":0,"history_lines":1,"events_offset":0,"events_lines":1,"console_offset":0,"console_lines":2,"uploaded_len":2}
+{"time":"2026-08-14T20:40:00.977965075Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:40:15.744587204Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":1,"history_lines":1,"events_offset":1,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-14T20:40:15.928506721Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:40:30.743869324Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":2,"history_lines":1,"events_offset":3,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-14T20:40:30.917560257Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:40:45.744229552Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":3,"history_lines":1,"events_offset":5,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-14T20:40:45.875347202Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:41:00.744908115Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":4,"history_lines":1,"events_offset":7,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-14T20:41:00.918429764Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:41:15.744313716Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":5,"history_lines":1,"events_offset":9,"events_lines":2,"console_offset":2,"console_lines":10}
+{"time":"2026-08-14T20:41:15.966956122Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:41:30.744909744Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":6,"history_lines":1,"events_offset":11,"events_lines":2,"console_offset":10,"console_lines":1}
+{"time":"2026-08-14T20:41:30.903737277Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:41:45.744811784Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":7,"history_lines":1,"events_offset":13,"events_lines":2,"console_offset":10,"console_lines":1}
+{"time":"2026-08-14T20:41:45.881876523Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:42:00.744201973Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":8,"history_lines":1,"events_offset":15,"events_lines":2,"console_offset":10,"console_lines":1}
+{"time":"2026-08-14T20:42:00.896435129Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:42:15.743927337Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":9,"history_lines":1,"events_offset":17,"events_lines":2,"console_offset":10,"console_lines":1}
+{"time":"2026-08-14T20:42:15.934416786Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:42:30.744737861Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":10,"history_lines":1,"events_offset":19,"events_lines":2,"console_offset":12,"console_lines":10}
+{"time":"2026-08-14T20:42:30.892376698Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:42:45.744822111Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":11,"history_lines":2,"events_offset":21,"events_lines":2,"console_offset":20,"console_lines":1}
+{"time":"2026-08-14T20:42:45.971531834Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:43:00.744614977Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":13,"history_lines":1,"events_offset":23,"events_lines":2,"console_offset":20,"console_lines":1}
+{"time":"2026-08-14T20:43:00.949080571Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:43:15.745651796Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":14,"history_lines":1,"events_offset":25,"events_lines":2,"console_offset":20,"console_lines":1}
+{"time":"2026-08-14T20:43:15.911365748Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:43:30.744257431Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":15,"history_lines":1,"events_offset":27,"events_lines":2,"console_offset":22,"console_lines":10}
+{"time":"2026-08-14T20:43:30.956144577Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}

zain/Activation/wandb/run-20260814_203945-j86czemo/logs/debug.log

@@ -0,0 +1,19 @@
+2026-08-14 20:39:45,019 INFO    MainThread:409188 [wandb_init.py:setup_run_log_directory():729] Logging user logs to /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/wandb/run-20260814_203945-j86czemo/logs/debug.log
+2026-08-14 20:39:45,019 INFO    MainThread:409188 [wandb_init.py:setup_run_log_directory():730] Logging internal logs to /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/wandb/run-20260814_203945-j86czemo/logs/debug-internal.log
+2026-08-14 20:39:45,019 INFO    MainThread:409188 [wandb_init.py:init():772] calling init triggers
+2026-08-14 20:39:45,019 INFO    MainThread:409188 [wandb_init.py:init():777] wandb.init called with sweep_config: {}
+config: {'_wandb': {}}
+2026-08-14 20:39:45,019 INFO    MainThread:409188 [wandb_init.py:init():820] starting backend
+2026-08-14 20:39:45,019 INFO    MainThread:409188 [wandb_init.py:init():835] sending inform_init request
+2026-08-14 20:39:45,279 INFO    MainThread:409188 [wandb_init.py:init():840] backend started and connected
+2026-08-14 20:39:45,280 INFO    MainThread:409188 [wandb_init.py:init():910] updated telemetry
+2026-08-14 20:39:45,287 INFO    MainThread:409188 [wandb_init.py:init():933] communicating run to backend with 90.0 second timeout
+2026-08-14 20:39:45,662 INFO    MainThread:409188 [wandb_init.py:init():978] starting run threads in backend
+2026-08-14 20:39:45,738 INFO    MainThread:409188 [wandb_run.py:_console_start():2621] atexit reg
+2026-08-14 20:39:45,738 INFO    MainThread:409188 [wandb_run.py:_redirect():2471] redirect: wrap_raw
+2026-08-14 20:39:45,738 INFO    MainThread:409188 [wandb_run.py:_redirect():2540] Wrapping output streams.
+2026-08-14 20:39:45,738 INFO    MainThread:409188 [wandb_run.py:_redirect():2563] Redirects installed.
+2026-08-14 20:39:45,739 INFO    MainThread:409188 [wandb_init.py:init():1016] run started, returning control to user process
+2026-08-14 20:39:45,740 INFO    MainThread:409188 [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': 9, '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-waleed10-9L_run', 'per_device_train_batch_size': 512, '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': 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-mlp-silu-waleed10-9L-2.0M-20260814-203944', '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': 512, '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-waleed10-9L', '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 20:39:45,742 INFO    MainThread:409188 [wandb_config.py:__setitem__():155] [no run ID] config set model/num_parameters = 2001280 - <bound method Run._config_callback of <wandb.sdk.wandb_run.Run object at 0x153f4a14ab50>>
+2026-08-14 20:39:45,742 INFO    MainThread:409188 [wandb_run.py:_config_callback():1346] config_cb model/num_parameters 2001280 None

zain/Activation/wandb/run-20260814_203945-j86czemo/run-j86czemo.wandb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dfed7a646bae5674ccf4bbb56d1c9fc949ee090944b63376022e56022a78c463
+size 163840