Auto upload zain 2026-08-12T20:27:58.442331

zain/Activation/out/glu-situglu-150L_run/checkpoint-1000/config.json

@@ -0,0 +1,36 @@
+{
+  "activation": "situglu",
+  "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": "glu",
+  "model_type": "tiny_llama",
+  "num_attention_heads": 4,
+  "num_hidden_layers": 150,
+  "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/glu-situglu-150L_run/checkpoint-1000/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:98db5f546f050aeeef120e86ee1d3d15dd97ff0683c69d9a4e272388b7951148
+size 50427120

zain/Activation/out/glu-situglu-150L_run/checkpoint-1000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:839c1a066f60a3b87f6a364403fc6e1a75dd7eda2ef03fa2a377ee59fdb1c757
+size 101708778

zain/Activation/out/glu-situglu-150L_run/checkpoint-1000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:645b71843375baf4cc10bc75a2f0f04e91b8e3f8e8929f518a87022898b3bc20
+size 14244

zain/Activation/out/glu-situglu-150L_run/checkpoint-1000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9db76f238494a413266398860b344c6c4c0f3a045ef78b23af1afac6775a12d1
+size 1064

zain/Activation/out/glu-situglu-150L_run/checkpoint-1000/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/glu-situglu-150L_run/checkpoint-1000/trainer_state.json

@@ -0,0 +1,544 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.06740815638692282,
+  "eval_steps": 50,
+  "global_step": 1000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0013481631277384564,
+      "grad_norm": 1.2578125,
+      "learning_rate": 0.000475,
+      "loss": 7.965972137451172,
+      "step": 20
+    },
+    {
+      "epoch": 0.002696326255476913,
+      "grad_norm": 0.875,
+      "learning_rate": 0.0009750000000000001,
+      "loss": 6.834014129638672,
+      "step": 40
+    },
+    {
+      "epoch": 0.003370407819346141,
+      "eval_loss": 6.037858009338379,
+      "eval_runtime": 23.5444,
+      "eval_samples_per_second": 404.639,
+      "eval_steps_per_second": 3.185,
+      "step": 50
+    },
+    {
+      "epoch": 0.004044489383215369,
+      "grad_norm": 0.2294921875,
+      "learning_rate": 0.001475,
+      "loss": 6.078770446777344,
+      "step": 60
+    },
+    {
+      "epoch": 0.005392652510953826,
+      "grad_norm": 0.119140625,
+      "learning_rate": 0.001975,
+      "loss": 6.011536407470703,
+      "step": 80
+    },
+    {
+      "epoch": 0.006740815638692282,
+      "grad_norm": 2.703125,
+      "learning_rate": 0.002475,
+      "loss": 5.908745956420899,
+      "step": 100
+    },
+    {
+      "epoch": 0.006740815638692282,
+      "eval_loss": 5.7961578369140625,
+      "eval_runtime": 23.4558,
+      "eval_samples_per_second": 406.169,
+      "eval_steps_per_second": 3.198,
+      "step": 100
+    },
+    {
+      "epoch": 0.008088978766430738,
+      "grad_norm": 0.1259765625,
+      "learning_rate": 0.0029749999999999998,
+      "loss": 5.797569274902344,
+      "step": 120
+    },
+    {
+      "epoch": 0.009437141894169195,
+      "grad_norm": 0.1953125,
+      "learning_rate": 0.003475,
+      "loss": 5.787290573120117,
+      "step": 140
+    },
+    {
+      "epoch": 0.010111223458038422,
+      "eval_loss": 5.742976188659668,
+      "eval_runtime": 23.5466,
+      "eval_samples_per_second": 404.602,
+      "eval_steps_per_second": 3.185,
+      "step": 150
+    },
+    {
+      "epoch": 0.010785305021907651,
+      "grad_norm": 0.52734375,
+      "learning_rate": 0.003975,
+      "loss": 5.71319808959961,
+      "step": 160
+    },
+    {
+      "epoch": 0.012133468149646108,
+      "grad_norm": 0.341796875,
+      "learning_rate": 0.004475,
+      "loss": 5.560100173950195,
+      "step": 180
+    },
+    {
+      "epoch": 0.013481631277384564,
+      "grad_norm": 0.443359375,
+      "learning_rate": 0.004975,
+      "loss": 5.274914169311524,
+      "step": 200
+    },
+    {
+      "epoch": 0.013481631277384564,
+      "eval_loss": 5.182697772979736,
+      "eval_runtime": 23.4825,
+      "eval_samples_per_second": 405.706,
+      "eval_steps_per_second": 3.194,
+      "step": 200
+    },
+    {
+      "epoch": 0.01482979440512302,
+      "grad_norm": 0.4140625,
+      "learning_rate": 0.005,
+      "loss": 5.2307487487792965,
+      "step": 220
+    },
+    {
+      "epoch": 0.016177957532861477,
+      "grad_norm": 1.15625,
+      "learning_rate": 0.005,
+      "loss": 4.977072906494141,
+      "step": 240
+    },
+    {
+      "epoch": 0.016852039096730706,
+      "eval_loss": 4.90332555770874,
+      "eval_runtime": 23.6215,
+      "eval_samples_per_second": 403.319,
+      "eval_steps_per_second": 3.175,
+      "step": 250
+    },
+    {
+      "epoch": 0.01752612066059993,
+      "grad_norm": 0.33203125,
+      "learning_rate": 0.005,
+      "loss": 4.896346664428711,
+      "step": 260
+    },
+    {
+      "epoch": 0.01887428378833839,
+      "grad_norm": 0.2373046875,
+      "learning_rate": 0.005,
+      "loss": 4.677676010131836,
+      "step": 280
+    },
+    {
+      "epoch": 0.020222446916076844,
+      "grad_norm": 0.66015625,
+      "learning_rate": 0.005,
+      "loss": 4.542130661010742,
+      "step": 300
+    },
+    {
+      "epoch": 0.020222446916076844,
+      "eval_loss": 4.47410249710083,
+      "eval_runtime": 23.5352,
+      "eval_samples_per_second": 404.797,
+      "eval_steps_per_second": 3.187,
+      "step": 300
+    },
+    {
+      "epoch": 0.021570610043815303,
+      "grad_norm": 0.44140625,
+      "learning_rate": 0.005,
+      "loss": 4.400739288330078,
+      "step": 320
+    },
+    {
+      "epoch": 0.022918773171553757,
+      "grad_norm": 0.35546875,
+      "learning_rate": 0.005,
+      "loss": 4.28521728515625,
+      "step": 340
+    },
+    {
+      "epoch": 0.023592854735422986,
+      "eval_loss": 4.220745086669922,
+      "eval_runtime": 23.526,
+      "eval_samples_per_second": 404.956,
+      "eval_steps_per_second": 3.188,
+      "step": 350
+    },
+    {
+      "epoch": 0.024266936299292215,
+      "grad_norm": 0.58203125,
+      "learning_rate": 0.005,
+      "loss": 4.232459259033203,
+      "step": 360
+    },
+    {
+      "epoch": 0.02561509942703067,
+      "grad_norm": 0.25,
+      "learning_rate": 0.005,
+      "loss": 4.133395767211914,
+      "step": 380
+    },
+    {
+      "epoch": 0.026963262554769128,
+      "grad_norm": 0.314453125,
+      "learning_rate": 0.005,
+      "loss": 4.02033576965332,
+      "step": 400
+    },
+    {
+      "epoch": 0.026963262554769128,
+      "eval_loss": 3.973088026046753,
+      "eval_runtime": 23.5308,
+      "eval_samples_per_second": 404.873,
+      "eval_steps_per_second": 3.187,
+      "step": 400
+    },
+    {
+      "epoch": 0.028311425682507583,
+      "grad_norm": 0.341796875,
+      "learning_rate": 0.005,
+      "loss": 3.916847991943359,
+      "step": 420
+    },
+    {
+      "epoch": 0.02965958881024604,
+      "grad_norm": 0.353515625,
+      "learning_rate": 0.005,
+      "loss": 3.8748004913330076,
+      "step": 440
+    },
+    {
+      "epoch": 0.030333670374115267,
+      "eval_loss": 3.780792474746704,
+      "eval_runtime": 23.4753,
+      "eval_samples_per_second": 405.83,
+      "eval_steps_per_second": 3.195,
+      "step": 450
+    },
+    {
+      "epoch": 0.031007751937984496,
+      "grad_norm": 0.2734375,
+      "learning_rate": 0.005,
+      "loss": 3.775615692138672,
+      "step": 460
+    },
+    {
+      "epoch": 0.032355915065722954,
+      "grad_norm": 0.396484375,
+      "learning_rate": 0.005,
+      "loss": 3.7166507720947264,
+      "step": 480
+    },
+    {
+      "epoch": 0.03370407819346141,
+      "grad_norm": 0.353515625,
+      "learning_rate": 0.005,
+      "loss": 3.6596424102783205,
+      "step": 500
+    },
+    {
+      "epoch": 0.03370407819346141,
+      "eval_loss": 3.6400845050811768,
+      "eval_runtime": 23.5481,
+      "eval_samples_per_second": 404.577,
+      "eval_steps_per_second": 3.185,
+      "step": 500
+    },
+    {
+      "epoch": 0.03505224132119986,
+      "grad_norm": 2.296875,
+      "learning_rate": 0.005,
+      "loss": 3.6132450103759766,
+      "step": 520
+    },
+    {
+      "epoch": 0.03640040444893832,
+      "grad_norm": 0.265625,
+      "learning_rate": 0.005,
+      "loss": 3.7193408966064454,
+      "step": 540
+    },
+    {
+      "epoch": 0.03707448601280755,
+      "eval_loss": 3.5622966289520264,
+      "eval_runtime": 23.4588,
+      "eval_samples_per_second": 406.117,
+      "eval_steps_per_second": 3.197,
+      "step": 550
+    },
+    {
+      "epoch": 0.03774856757667678,
+      "grad_norm": 0.244140625,
+      "learning_rate": 0.005,
+      "loss": 3.5610897064208986,
+      "step": 560
+    },
+    {
+      "epoch": 0.03909673070441524,
+      "grad_norm": 0.2294921875,
+      "learning_rate": 0.005,
+      "loss": 3.4865718841552735,
+      "step": 580
+    },
+    {
+      "epoch": 0.04044489383215369,
+      "grad_norm": 0.333984375,
+      "learning_rate": 0.005,
+      "loss": 3.467424774169922,
+      "step": 600
+    },
+    {
+      "epoch": 0.04044489383215369,
+      "eval_loss": 3.442713737487793,
+      "eval_runtime": 23.5927,
+      "eval_samples_per_second": 403.812,
+      "eval_steps_per_second": 3.179,
+      "step": 600
+    },
+    {
+      "epoch": 0.04179305695989215,
+      "grad_norm": 0.294921875,
+      "learning_rate": 0.005,
+      "loss": 3.4243141174316407,
+      "step": 620
+    },
+    {
+      "epoch": 0.043141220087630605,
+      "grad_norm": 0.26171875,
+      "learning_rate": 0.005,
+      "loss": 3.4109996795654296,
+      "step": 640
+    },
+    {
+      "epoch": 0.043815301651499834,
+      "eval_loss": 3.3643105030059814,
+      "eval_runtime": 23.5407,
+      "eval_samples_per_second": 404.703,
+      "eval_steps_per_second": 3.186,
+      "step": 650
+    },
+    {
+      "epoch": 0.044489383215369056,
+      "grad_norm": 0.2255859375,
+      "learning_rate": 0.005,
+      "loss": 3.35318603515625,
+      "step": 660
+    },
+    {
+      "epoch": 0.045837546343107514,
+      "grad_norm": 0.2490234375,
+      "learning_rate": 0.005,
+      "loss": 3.3295913696289063,
+      "step": 680
+    },
+    {
+      "epoch": 0.04718570947084597,
+      "grad_norm": 0.23046875,
+      "learning_rate": 0.005,
+      "loss": 3.2915592193603516,
+      "step": 700
+    },
+    {
+      "epoch": 0.04718570947084597,
+      "eval_loss": 3.2848241329193115,
+      "eval_runtime": 23.4573,
+      "eval_samples_per_second": 406.142,
+      "eval_steps_per_second": 3.197,
+      "step": 700
+    },
+    {
+      "epoch": 0.04853387259858443,
+      "grad_norm": 0.318359375,
+      "learning_rate": 0.005,
+      "loss": 3.2676395416259765,
+      "step": 720
+    },
+    {
+      "epoch": 0.04988203572632288,
+      "grad_norm": 0.265625,
+      "learning_rate": 0.005,
+      "loss": 3.2363258361816407,
+      "step": 740
+    },
+    {
+      "epoch": 0.05055611729019211,
+      "eval_loss": 3.2303872108459473,
+      "eval_runtime": 23.514,
+      "eval_samples_per_second": 405.162,
+      "eval_steps_per_second": 3.19,
+      "step": 750
+    },
+    {
+      "epoch": 0.05123019885406134,
+      "grad_norm": 0.244140625,
+      "learning_rate": 0.005,
+      "loss": 3.2281532287597656,
+      "step": 760
+    },
+    {
+      "epoch": 0.0525783619817998,
+      "grad_norm": 0.26171875,
+      "learning_rate": 0.005,
+      "loss": 3.227946472167969,
+      "step": 780
+    },
+    {
+      "epoch": 0.053926525109538256,
+      "grad_norm": 0.333984375,
+      "learning_rate": 0.005,
+      "loss": 3.187031555175781,
+      "step": 800
+    },
+    {
+      "epoch": 0.053926525109538256,
+      "eval_loss": 3.1891424655914307,
+      "eval_runtime": 23.6088,
+      "eval_samples_per_second": 403.536,
+      "eval_steps_per_second": 3.177,
+      "step": 800
+    },
+    {
+      "epoch": 0.05527468823727671,
+      "grad_norm": 0.2451171875,
+      "learning_rate": 0.005,
+      "loss": 3.181216239929199,
+      "step": 820
+    },
+    {
+      "epoch": 0.056622851365015166,
+      "grad_norm": 0.291015625,
+      "learning_rate": 0.005,
+      "loss": 3.157722282409668,
+      "step": 840
+    },
+    {
+      "epoch": 0.057296932928884395,
+      "eval_loss": 3.1340532302856445,
+      "eval_runtime": 23.5359,
+      "eval_samples_per_second": 404.786,
+      "eval_steps_per_second": 3.187,
+      "step": 850
+    },
+    {
+      "epoch": 0.057971014492753624,
+      "grad_norm": 0.244140625,
+      "learning_rate": 0.005,
+      "loss": 3.124161720275879,
+      "step": 860
+    },
+    {
+      "epoch": 0.05931917762049208,
+      "grad_norm": 0.26953125,
+      "learning_rate": 0.005,
+      "loss": 3.1029438018798827,
+      "step": 880
+    },
+    {
+      "epoch": 0.06066734074823053,
+      "grad_norm": 0.302734375,
+      "learning_rate": 0.005,
+      "loss": 3.0970760345458985,
+      "step": 900
+    },
+    {
+      "epoch": 0.06066734074823053,
+      "eval_loss": 3.0905423164367676,
+      "eval_runtime": 23.515,
+      "eval_samples_per_second": 405.145,
+      "eval_steps_per_second": 3.189,
+      "step": 900
+    },
+    {
+      "epoch": 0.06201550387596899,
+      "grad_norm": 0.3515625,
+      "learning_rate": 0.005,
+      "loss": 3.076877784729004,
+      "step": 920
+    },
+    {
+      "epoch": 0.06336366700370745,
+      "grad_norm": 0.306640625,
+      "learning_rate": 0.005,
+      "loss": 3.0760448455810545,
+      "step": 940
+    },
+    {
+      "epoch": 0.06403774856757667,
+      "eval_loss": 3.056104898452759,
+      "eval_runtime": 23.5713,
+      "eval_samples_per_second": 404.178,
+      "eval_steps_per_second": 3.182,
+      "step": 950
+    },
+    {
+      "epoch": 0.06471183013144591,
+      "grad_norm": 0.2333984375,
+      "learning_rate": 0.005,
+      "loss": 3.046922492980957,
+      "step": 960
+    },
+    {
+      "epoch": 0.06605999325918437,
+      "grad_norm": 0.255859375,
+      "learning_rate": 0.005,
+      "loss": 3.026295471191406,
+      "step": 980
+    },
+    {
+      "epoch": 0.06740815638692282,
+      "grad_norm": 0.271484375,
+      "learning_rate": 0.005,
+      "loss": 3.0347896575927735,
+      "step": 1000
+    },
+    {
+      "epoch": 0.06740815638692282,
+      "eval_loss": 3.0281519889831543,
+      "eval_runtime": 23.6926,
+      "eval_samples_per_second": 402.109,
+      "eval_steps_per_second": 3.166,
+      "step": 1000
+    }
+  ],
+  "logging_steps": 20,
+  "max_steps": 1000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 1000,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4839413121024000.0,
+  "train_batch_size": 64,
+  "trial_name": null,
+  "trial_params": null
+}

zain/Activation/out/glu-situglu-150L_run/checkpoint-1000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:911a0efdde4eddfa1b9b0003a7a4193524045592e5238e63d365aab81e6e0cac
+size 4920

zain/Activation/out/glu-situglu-150L_run/config.json

@@ -0,0 +1,36 @@
+{
+  "activation": "situglu",
+  "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": "glu",
+  "model_type": "tiny_llama",
+  "num_attention_heads": 4,
+  "num_hidden_layers": 150,
+  "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/glu-situglu-150L_run/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:98db5f546f050aeeef120e86ee1d3d15dd97ff0683c69d9a4e272388b7951148
+size 50427120

zain/Activation/out/glu-situglu-150L_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/glu-situglu-150L_run/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:911a0efdde4eddfa1b9b0003a7a4193524045592e5238e63d365aab81e6e0cac
+size 4920

zain/Activation/out/glu-situglu-150L_run/training_log.jsonl

@@ -60,3 +60,13 @@
 {"step": 860, "epoch": 0.057971014492753624, "timestamp": 1786566140.4038272, "loss": 3.124161720275879, "grad_norm": 0.244140625, "learning_rate": 0.005, "train/total_time_seconds": 339.5495206862688, "train/time_per_step_avg": 0.3946998628601432, "train/epoch_time_elapsed": 782.8972343765199, "train/estimated_remaining_minutes": 0.9212583894588688}
 {"step": 880, "epoch": 0.05931917762049208, "timestamp": 1786566149.3396704, "loss": 3.1029438018798827, "grad_norm": 0.26953125, "learning_rate": 0.005, "train/total_time_seconds": 347.4950982891023, "train/time_per_step_avg": 0.3962212897092104, "train/epoch_time_elapsed": 791.8330778218806, "train/estimated_remaining_minutes": 0.7897615870206871}
 {"step": 900, "epoch": 0.06066734074823053, "timestamp": 1786566158.2259598, "loss": 3.0970760345458985, "grad_norm": 0.302734375, "learning_rate": 0.005, "train/total_time_seconds": 355.3894107863307, "train/time_per_step_avg": 0.39733249008655547, "train/epoch_time_elapsed": 800.719366915524, "train/estimated_remaining_minutes": 0.658128538493205}
+{"step": 900, "epoch": 0.06066734074823053, "timestamp": 1786566181.7424457, "eval_loss": 3.0905423164367676, "eval_runtime": 23.515, "eval_samples_per_second": 405.145, "eval_steps_per_second": 3.189, "train/total_time_seconds": 355.3894107863307, "train/time_per_step_avg": 0.39733249008655547, "train/epoch_time_elapsed": 824.2358520179987, "train/estimated_remaining_minutes": 0.658128538493205}
+{"step": 920, "epoch": 0.06201550387596899, "timestamp": 1786566190.6065702, "loss": 3.076877784729004, "grad_norm": 0.3515625, "learning_rate": 0.005, "train/total_time_seconds": 363.2537060081959, "train/time_per_step_avg": 0.39502713322639466, "train/epoch_time_elapsed": 833.0999770723283, "train/estimated_remaining_minutes": 0.5264546463886897}
+{"step": 940, "epoch": 0.06336366700370745, "timestamp": 1786566199.5078952, "loss": 3.0760448455810545, "grad_norm": 0.306640625, "learning_rate": 0.005, "train/total_time_seconds": 371.1556162238121, "train/time_per_step_avg": 0.395236103720963, "train/epoch_time_elapsed": 842.0013028644025, "train/estimated_remaining_minutes": 0.39484640023809797}
+{"step": 950, "epoch": 0.06403774856757667, "timestamp": 1786566227.743793, "eval_loss": 3.056104898452759, "eval_runtime": 23.5713, "eval_samples_per_second": 404.178, "eval_steps_per_second": 3.182, "train/total_time_seconds": 375.3206667341292, "train/time_per_step_avg": 0.3971567253395915, "train/epoch_time_elapsed": 870.2371992990375, "train/estimated_remaining_minutes": 0.32922865502993787}
+{"step": 960, "epoch": 0.06471183013144591, "timestamp": 1786566232.1631339, "loss": 3.046922492980957, "grad_norm": 0.2333984375, "learning_rate": 0.005, "train/total_time_seconds": 379.24332094192505, "train/time_per_step_avg": 0.3969380025565624, "train/epoch_time_elapsed": 874.6565411239862, "train/estimated_remaining_minutes": 0.2633634173207813}
+{"step": 980, "epoch": 0.06605999325918437, "timestamp": 1786566241.2370214, "loss": 3.026295471191406, "grad_norm": 0.255859375, "learning_rate": 0.005, "train/total_time_seconds": 387.2918460480869, "train/time_per_step_avg": 0.39796747758984563, "train/epoch_time_elapsed": 883.7304284498096, "train/estimated_remaining_minutes": 0.13173192042451937}
+{"step": 1000, "epoch": 0.06740815638692282, "timestamp": 1786566250.13924, "loss": 3.0347896575927735, "grad_norm": 0.271484375, "learning_rate": 0.005, "train/total_time_seconds": 395.203328166157, "train/time_per_step_avg": 0.3981391737982631, "train/epoch_time_elapsed": 892.632646843791, "train/estimated_remaining_minutes": 0.0}
+{"step": 1000, "epoch": 0.06740815638692282, "timestamp": 1786566273.8332717, "eval_loss": 3.0281519889831543, "eval_runtime": 23.6926, "eval_samples_per_second": 402.109, "eval_steps_per_second": 3.166, "train/total_time_seconds": 395.203328166157, "train/time_per_step_avg": 0.3981391737982631, "train/epoch_time_elapsed": 916.3266776278615, "train/estimated_remaining_minutes": 0.0}
+{"step": 1000, "epoch": 0.06740815638692282, "timestamp": 1786566274.2843466, "train_runtime": 917.6128, "train_samples_per_second": 69.746, "train_steps_per_second": 1.09, "total_flos": 4839413121024000.0, "train_loss": 4.140005214691162, "train/total_time_seconds": 395.203328166157, "train/time_per_step_avg": 0.3981391737982631, "train/epoch_time_elapsed": 916.7777519077063, "train/estimated_remaining_minutes": 0.0}
+{"step": 1000, "epoch": 0.06740815638692282, "timestamp": 1786566297.7726266, "eval_loss": 3.0281519889831543, "eval_runtime": 23.485, "eval_samples_per_second": 405.664, "eval_steps_per_second": 3.194, "train/total_time_seconds": 395.203328166157, "train/time_per_step_avg": 0.3981391737982631, "train/epoch_time_elapsed": 940.2660321295261, "train/estimated_remaining_minutes": 0.0}

zain/Activation/out/glu-waleed-150L_run/training_log.jsonl

@@ -0,0 +1,14 @@
+{"step": 20, "epoch": 0.0013481631277384564, "timestamp": 1786566309.6015754, "loss": 7.956727600097656, "grad_norm": 1.2578125, "learning_rate": 0.000475, "train/total_time_seconds": 7.922187950462103, "train/time_per_step_avg": 0.39610939752310514, "train/epoch_time_elapsed": 9.054701536893845, "train/estimated_remaining_minutes": 6.469786826210718}
+{"step": 40, "epoch": 0.002696326255476913, "timestamp": 1786566318.070938, "loss": 6.816352844238281, "grad_norm": 0.8671875, "learning_rate": 0.0009750000000000001, "train/total_time_seconds": 15.391319837421179, "train/time_per_step_avg": 0.38478299593552945, "train/epoch_time_elapsed": 17.524064287543297, "train/estimated_remaining_minutes": 6.156527934968471}
+{"step": 50, "epoch": 0.003370407819346141, "timestamp": 1786566345.1053584, "eval_loss": 6.0367817878723145, "eval_runtime": 22.7925, "eval_samples_per_second": 417.988, "eval_steps_per_second": 3.291, "train/total_time_seconds": 19.131311435252428, "train/time_per_step_avg": 0.3826262287050486, "train/epoch_time_elapsed": 44.55848413333297, "train/estimated_remaining_minutes": 6.058248621163269}
+{"step": 60, "epoch": 0.004044489383215369, "timestamp": 1786566349.3442862, "loss": 6.0750377655029295, "grad_norm": 0.2431640625, "learning_rate": 0.001475, "train/total_time_seconds": 22.86884220317006, "train/time_per_step_avg": 0.38114737005283433, "train/epoch_time_elapsed": 48.79741337522864, "train/estimated_remaining_minutes": 5.971308797494404}
+{"step": 80, "epoch": 0.005392652510953826, "timestamp": 1786566357.8223898, "loss": 5.994783401489258, "grad_norm": 0.68359375, "learning_rate": 0.001975, "train/total_time_seconds": 30.345024030655622, "train/time_per_step_avg": 0.37931280038319526, "train/epoch_time_elapsed": 57.27551680803299, "train/estimated_remaining_minutes": 5.816129605875661}
+{"step": 100, "epoch": 0.006740815638692282, "timestamp": 1786566366.4805279, "loss": 5.948408126831055, "grad_norm": 4.09375, "learning_rate": 0.002475, "train/total_time_seconds": 37.975685119628906, "train/time_per_step_avg": 0.37975685119628905, "train/epoch_time_elapsed": 65.93365533649921, "train/estimated_remaining_minutes": 5.696352767944336}
+{"step": 100, "epoch": 0.006740815638692282, "timestamp": 1786566389.4252908, "eval_loss": 5.804104804992676, "eval_runtime": 22.9432, "eval_samples_per_second": 415.243, "eval_steps_per_second": 3.269, "train/total_time_seconds": 37.975685119628906, "train/time_per_step_avg": 0.37975685119628905, "train/epoch_time_elapsed": 88.8784164711833, "train/estimated_remaining_minutes": 5.696352767944336}
+{"step": 120, "epoch": 0.008088978766430738, "timestamp": 1786566398.227304, "loss": 5.760046768188476, "grad_norm": 0.1318359375, "learning_rate": 0.0029749999999999998, "train/total_time_seconds": 45.74835725128651, "train/time_per_step_avg": 0.378261693008244, "train/epoch_time_elapsed": 97.68043184280396, "train/estimated_remaining_minutes": 5.591465886268351}
+{"step": 140, "epoch": 0.009437141894169195, "timestamp": 1786566407.0935357, "loss": 5.92535171508789, "grad_norm": 0.294921875, "learning_rate": 0.003475, "train/total_time_seconds": 53.6095660738647, "train/time_per_step_avg": 0.3821824623644352, "train/epoch_time_elapsed": 106.54666317626834, "train/estimated_remaining_minutes": 5.488598431371861}
+{"step": 150, "epoch": 0.010111223458038422, "timestamp": 1786566434.3153608, "eval_loss": 6.016007423400879, "eval_runtime": 22.9277, "eval_samples_per_second": 415.523, "eval_steps_per_second": 3.271, "train/total_time_seconds": 57.40382784232497, "train/time_per_step_avg": 0.38272516407072543, "train/epoch_time_elapsed": 133.7684867233038, "train/estimated_remaining_minutes": 5.421472629552914}
+{"step": 160, "epoch": 0.010785305021907651, "timestamp": 1786566438.590914, "loss": 5.983197021484375, "grad_norm": 0.357421875, "learning_rate": 0.003975, "train/total_time_seconds": 61.17868063226342, "train/time_per_step_avg": 0.3830983842909336, "train/epoch_time_elapsed": 138.0440418124199, "train/estimated_remaining_minutes": 5.35313455532305}
+{"step": 180, "epoch": 0.012133468149646108, "timestamp": 1786566447.1577578, "loss": 5.803388977050782, "grad_norm": 0.546875, "learning_rate": 0.004475, "train/total_time_seconds": 68.74973602592945, "train/time_per_step_avg": 0.38404711995273827, "train/epoch_time_elapsed": 146.61088536307216, "train/estimated_remaining_minutes": 5.21988736493168}
+{"step": 200, "epoch": 0.013481631277384564, "timestamp": 1786566455.7456908, "loss": 5.976637649536133, "grad_norm": 0.1611328125, "learning_rate": 0.004975, "train/total_time_seconds": 76.32484335824847, "train/time_per_step_avg": 0.3834915823861957, "train/epoch_time_elapsed": 155.1988184414804, "train/estimated_remaining_minutes": 5.0883228905498985}
+{"step": 200, "epoch": 0.013481631277384564, "timestamp": 1786566478.5184085, "eval_loss": 5.777098178863525, "eval_runtime": 22.7713, "eval_samples_per_second": 418.378, "eval_steps_per_second": 3.294, "train/total_time_seconds": 76.32484335824847, "train/time_per_step_avg": 0.3834915823861957, "train/epoch_time_elapsed": 177.97153493389487, "train/estimated_remaining_minutes": 5.0883228905498985}

zain/Activation/wandb/debug-internal.log

@@ -1,117 +1,31 @@
-{"time":"2026-08-12T20:09:16.738503108Z","level":"INFO","msg":"wandb-core"}
-{"time":"2026-08-12T20:09:16.738638331Z","level":"INFO","msg":"stream: starting","core version":"0.28.1"}
-{"time":"2026-08-12T20:09:16.997898196Z","level":"INFO","msg":"stream: created new stream","id":"iwodu993"}
-{"time":"2026-08-12T20:09:16.997998486Z","level":"INFO","msg":"handler: started"}
-{"time":"2026-08-12T20:09:16.998077479Z","level":"INFO","msg":"stream: started"}
-{"time":"2026-08-12T20:09:16.998089219Z","level":"INFO","msg":"writer: started","stream_id":"iwodu993"}
-{"time":"2026-08-12T20:09:16.998115578Z","level":"INFO","msg":"sender: started"}
-{"time":"2026-08-12T20:09:18.44005441Z","level":"INFO","msg":"filestream: sending request","total_files":1,"console_offset":0,"console_lines":1}
-{"time":"2026-08-12T20:09:18.532368989Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:09:33.440564239Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":0,"history_lines":1,"events_offset":0,"events_lines":2,"console_offset":0,"console_lines":2,"uploaded_len":2}
-{"time":"2026-08-12T20:09:33.563513759Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:09:48.441045184Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":1,"history_lines":1,"events_offset":2,"events_lines":2,"console_offset":0,"console_lines":1}
-{"time":"2026-08-12T20:09:48.557576002Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:10:03.440654516Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":4,"events_lines":2,"console_offset":2,"console_lines":2}
-{"time":"2026-08-12T20:10:03.572399602Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:10:18.440716882Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":2,"history_lines":3,"events_offset":6,"events_lines":2,"console_offset":0,"console_lines":1}
-{"time":"2026-08-12T20:10:18.570821459Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:10:33.44025169Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":5,"history_lines":1,"events_offset":8,"events_lines":2,"console_offset":0,"console_lines":1}
-{"time":"2026-08-12T20:10:33.578744088Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:10:48.44111971Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":10,"events_lines":2,"console_offset":3,"console_lines":5}
-{"time":"2026-08-12T20:10:48.553409185Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:11:03.440521667Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":6,"history_lines":2,"events_offset":12,"events_lines":2,"console_offset":0,"console_lines":1}
-{"time":"2026-08-12T20:11:03.584243299Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:11:18.440850458Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":8,"history_lines":1,"events_offset":14,"events_lines":2,"console_offset":0,"console_lines":1}
-{"time":"2026-08-12T20:11:18.56141202Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:11:33.440539856Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":16,"events_lines":2,"console_offset":3,"console_lines":1}
-{"time":"2026-08-12T20:11:33.561771445Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:11:48.441041429Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":9,"history_lines":2,"events_offset":18,"events_lines":2,"console_offset":0,"console_lines":1}
-{"time":"2026-08-12T20:11:48.545380786Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:12:03.441043559Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":11,"history_lines":2,"events_offset":20,"events_lines":2,"console_offset":0,"console_lines":1}
-{"time":"2026-08-12T20:12:03.628655493Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:12:18.440969014Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":22,"events_lines":2,"console_offset":3,"console_lines":1}
-{"time":"2026-08-12T20:12:18.5610872Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:12:33.441124864Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":13,"history_lines":2,"events_offset":24,"events_lines":2,"console_offset":0,"console_lines":1}
-{"time":"2026-08-12T20:12:33.57030471Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:12:48.440666824Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":15,"history_lines":1,"events_offset":26,"events_lines":2,"console_offset":0,"console_lines":1}
-{"time":"2026-08-12T20:12:48.565353887Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:13:03.440576164Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":28,"events_lines":2,"console_offset":3,"console_lines":1}
-{"time":"2026-08-12T20:13:03.579753663Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:13:18.440547728Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":16,"history_lines":2,"events_offset":30,"events_lines":2,"console_offset":0,"console_lines":1}
-{"time":"2026-08-12T20:13:18.564860093Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:13:33.440592385Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":18,"history_lines":2,"events_offset":32,"events_lines":2,"console_offset":0,"console_lines":1}
-{"time":"2026-08-12T20:13:33.548485571Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:13:48.44016866Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":34,"events_lines":2,"console_offset":3,"console_lines":1}
-{"time":"2026-08-12T20:13:48.583030594Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:14:03.440544008Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":20,"history_lines":2,"events_offset":36,"events_lines":2,"console_offset":0,"console_lines":1}
-{"time":"2026-08-12T20:14:03.597324029Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:14:18.440390272Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":22,"history_lines":1,"events_offset":38,"events_lines":2,"console_offset":0,"console_lines":1}
-{"time":"2026-08-12T20:14:18.55788272Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:14:33.441059972Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":40,"events_lines":2,"console_offset":3,"console_lines":1}
-{"time":"2026-08-12T20:14:33.587036938Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:14:48.440802646Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":23,"history_lines":2,"events_offset":42,"events_lines":2,"console_offset":0,"console_lines":1}
-{"time":"2026-08-12T20:14:48.557474094Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:15:03.440209302Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":25,"history_lines":2,"events_offset":44,"events_lines":2,"console_offset":0,"console_lines":1}
-{"time":"2026-08-12T20:15:03.573843696Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:15:18.440472959Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":46,"events_lines":2,"console_offset":3,"console_lines":1}
-{"time":"2026-08-12T20:15:18.575346389Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:15:33.440665029Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":27,"history_lines":2,"events_offset":48,"events_lines":2,"console_offset":0,"console_lines":1}
-{"time":"2026-08-12T20:15:33.547841301Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:15:48.440945115Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":29,"history_lines":1,"events_offset":50,"events_lines":2,"console_offset":0,"console_lines":1}
-{"time":"2026-08-12T20:15:48.580689018Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:16:03.440387306Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":52,"events_lines":2,"console_offset":3,"console_lines":1}
-{"time":"2026-08-12T20:16:03.616854543Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:16:18.440973259Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":30,"history_lines":2,"events_offset":54,"events_lines":2,"console_offset":0,"console_lines":1}
-{"time":"2026-08-12T20:16:18.79807476Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:16:33.440227014Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":32,"history_lines":2,"events_offset":56,"events_lines":2,"console_offset":3,"console_lines":1}
-{"time":"2026-08-12T20:16:33.565290156Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:16:48.440742454Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":58,"events_lines":2,"console_offset":8,"console_lines":28}
-{"time":"2026-08-12T20:16:48.551189454Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:17:03.440956902Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":34,"history_lines":1,"events_offset":60,"events_lines":2,"console_offset":36,"console_lines":1}
-{"time":"2026-08-12T20:17:03.605196494Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:17:18.440586179Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":35,"history_lines":2,"events_offset":62,"events_lines":2,"console_offset":37,"console_lines":2}
-{"time":"2026-08-12T20:17:18.563680582Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:17:33.44094423Z","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":64,"events_lines":2}
-{"time":"2026-08-12T20:17:33.561996152Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:17:48.440242476Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":37,"history_lines":2,"events_offset":66,"events_lines":2,"console_offset":39,"console_lines":2}
-{"time":"2026-08-12T20:17:48.563852714Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:18:03.440710904Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":39,"history_lines":1,"events_offset":68,"events_lines":2,"console_offset":41,"console_lines":1}
-{"time":"2026-08-12T20:18:03.630833344Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:18:18.440467933Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":40,"history_lines":1,"events_offset":70,"events_lines":2,"console_offset":42,"console_lines":1}
-{"time":"2026-08-12T20:18:18.58762668Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:18:33.440624979Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":41,"history_lines":1,"events_offset":72,"events_lines":2,"console_offset":43,"console_lines":1}
-{"time":"2026-08-12T20:18:33.577661637Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:18:48.44083601Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":42,"history_lines":2,"events_offset":74,"events_lines":2,"console_offset":44,"console_lines":2}
-{"time":"2026-08-12T20:18:48.608976332Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:19:03.440624861Z","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":76,"events_lines":2}
-{"time":"2026-08-12T20:19:03.604950399Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:19:18.44257959Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":44,"history_lines":2,"events_offset":78,"events_lines":2,"console_offset":46,"console_lines":2}
-{"time":"2026-08-12T20:19:18.56511861Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:19:33.440886853Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":46,"history_lines":1,"events_offset":80,"events_lines":2,"console_offset":48,"console_lines":1}
-{"time":"2026-08-12T20:19:33.577525338Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:19:48.440385287Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":47,"history_lines":1,"events_offset":82,"events_lines":2,"console_offset":49,"console_lines":1}
-{"time":"2026-08-12T20:19:48.613403356Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:20:03.441010477Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":48,"history_lines":1,"events_offset":84,"events_lines":2,"console_offset":50,"console_lines":1}
-{"time":"2026-08-12T20:20:03.634061138Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:20:18.440400526Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":49,"history_lines":2,"events_offset":86,"events_lines":2,"console_offset":51,"console_lines":2}
-{"time":"2026-08-12T20:20:18.570016778Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:20:33.440891004Z","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":88,"events_lines":2}
-{"time":"2026-08-12T20:20:33.580087158Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:20:48.440972097Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":51,"history_lines":1,"events_offset":90,"events_lines":2,"console_offset":53,"console_lines":1}
-{"time":"2026-08-12T20:20:48.583660679Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:21:03.440841778Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":52,"history_lines":2,"events_offset":92,"events_lines":2,"console_offset":54,"console_lines":2}
-{"time":"2026-08-12T20:21:03.583536338Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:21:18.441191767Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":54,"history_lines":1,"events_offset":94,"events_lines":2,"console_offset":56,"console_lines":1}
-{"time":"2026-08-12T20:21:18.554031834Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:21:33.441098604Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":55,"history_lines":1,"events_offset":96,"events_lines":2,"console_offset":57,"console_lines":1}
-{"time":"2026-08-12T20:21:33.596505268Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:21:48.440653336Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":56,"history_lines":2,"events_offset":98,"events_lines":2,"console_offset":58,"console_lines":2}
-{"time":"2026-08-12T20:21:48.565447975Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:22:03.441099564Z","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":100,"events_lines":2}
-{"time":"2026-08-12T20:22:03.575982676Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:22:18.440837339Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":58,"history_lines":1,"events_offset":102,"events_lines":2,"console_offset":60,"console_lines":1}
-{"time":"2026-08-12T20:22:18.599874482Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:22:33.44122393Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":59,"history_lines":2,"events_offset":104,"events_lines":2,"console_offset":61,"console_lines":2}
-{"time":"2026-08-12T20:22:33.563163088Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-12T20:22:48.440357323Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":61,"history_lines":1,"events_offset":106,"events_lines":2,"console_offset":63,"console_lines":1}
-{"time":"2026-08-12T20:22:48.562864965Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T20:24:59.716522619Z","level":"INFO","msg":"wandb-core"}
+{"time":"2026-08-12T20:24:59.716668506Z","level":"INFO","msg":"stream: starting","core version":"0.28.1"}
+{"time":"2026-08-12T20:24:59.978902588Z","level":"INFO","msg":"stream: created new stream","id":"k14eff4o"}
+{"time":"2026-08-12T20:24:59.979002733Z","level":"INFO","msg":"handler: started"}
+{"time":"2026-08-12T20:24:59.97909183Z","level":"INFO","msg":"stream: started"}
+{"time":"2026-08-12T20:24:59.979109732Z","level":"INFO","msg":"writer: started","stream_id":"k14eff4o"}
+{"time":"2026-08-12T20:24:59.979127696Z","level":"INFO","msg":"sender: started"}
+{"time":"2026-08-12T20:25:01.477268978Z","level":"INFO","msg":"filestream: sending request","total_files":1,"console_offset":0,"console_lines":1}
+{"time":"2026-08-12T20:25:01.570884465Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T20:25:16.477745126Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":0,"history_lines":1,"events_offset":0,"events_lines":2,"console_offset":0,"console_lines":2,"uploaded_len":2}
+{"time":"2026-08-12T20:25:16.595722633Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T20:25:31.477966084Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":1,"history_lines":1,"events_offset":2,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-12T20:25:31.628536652Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T20:25:46.478040478Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":2,"history_lines":1,"events_offset":4,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-12T20:25:46.60071818Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T20:26:01.478205323Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":3,"history_lines":2,"events_offset":6,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-12T20:26:01.619749748Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T20:26:16.478436053Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":5,"history_lines":1,"events_offset":8,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-12T20:26:16.638029625Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T20:26:31.478304245Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":6,"history_lines":1,"events_offset":10,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-12T20:26:31.588473215Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T20:26:46.47769931Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":7,"history_lines":1,"events_offset":12,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-12T20:26:46.642698181Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T20:27:01.478376889Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":8,"history_lines":1,"events_offset":14,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-12T20:27:01.614026299Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T20:27:16.478276617Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":9,"history_lines":1,"events_offset":16,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-12T20:27:16.59518949Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T20:27:31.477617651Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":10,"history_lines":2,"events_offset":18,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-12T20:27:31.615253909Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-12T20:27:46.477671505Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":12,"history_lines":1,"events_offset":20,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-12T20:27:46.615859318Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}

zain/Activation/wandb/debug.log

@@ -1,20 +1,20 @@
-2026-08-12 20:09:16,737 INFO    MainThread:1371664 [wandb_init.py:setup_run_log_directory():729] Logging user logs to /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/wandb/run-20260812_200916-iwodu993/logs/debug.log
-2026-08-12 20:09:16,737 INFO    MainThread:1371664 [wandb_init.py:setup_run_log_directory():730] Logging internal logs to /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/wandb/run-20260812_200916-iwodu993/logs/debug-internal.log
-2026-08-12 20:09:16,737 INFO    MainThread:1371664 [wandb_init.py:init():772] calling init triggers
-2026-08-12 20:09:16,737 INFO    MainThread:1371664 [wandb_init.py:init():777] wandb.init called with sweep_config: {}
+2026-08-12 20:24:59,715 INFO    MainThread:1371664 [wandb_init.py:setup_run_log_directory():729] Logging user logs to /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/wandb/run-20260812_202459-k14eff4o/logs/debug.log
+2026-08-12 20:24:59,715 INFO    MainThread:1371664 [wandb_init.py:setup_run_log_directory():730] Logging internal logs to /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/wandb/run-20260812_202459-k14eff4o/logs/debug-internal.log
+2026-08-12 20:24:59,715 INFO    MainThread:1371664 [wandb_init.py:init():772] calling init triggers
+2026-08-12 20:24:59,715 INFO    MainThread:1371664 [wandb_init.py:init():777] wandb.init called with sweep_config: {}
 config: {'_wandb': {}}
-2026-08-12 20:09:16,737 INFO    MainThread:1371664 [wandb_init.py:init():820] starting backend
-2026-08-12 20:09:16,737 INFO    MainThread:1371664 [wandb_init.py:init():826] Connected to an existing wandb-core service via WANDB_SERVICE
-2026-08-12 20:09:16,737 INFO    MainThread:1371664 [wandb_init.py:init():835] sending inform_init request
-2026-08-12 20:09:16,998 INFO    MainThread:1371664 [wandb_init.py:init():840] backend started and connected
-2026-08-12 20:09:16,999 INFO    MainThread:1371664 [wandb_init.py:init():910] updated telemetry
-2026-08-12 20:09:17,006 INFO    MainThread:1371664 [wandb_init.py:init():933] communicating run to backend with 90.0 second timeout
-2026-08-12 20:09:17,417 INFO    MainThread:1371664 [wandb_init.py:init():978] starting run threads in backend
-2026-08-12 20:09:17,496 INFO    MainThread:1371664 [wandb_run.py:_console_start():2621] atexit reg
-2026-08-12 20:09:17,496 INFO    MainThread:1371664 [wandb_run.py:_redirect():2471] redirect: wrap_raw
-2026-08-12 20:09:17,496 INFO    MainThread:1371664 [wandb_run.py:_redirect():2540] Wrapping output streams.
-2026-08-12 20:09:17,496 INFO    MainThread:1371664 [wandb_run.py:_redirect():2563] Redirects installed.
-2026-08-12 20:09:17,497 INFO    MainThread:1371664 [wandb_init.py:init():1016] run started, returning control to user process
-2026-08-12 20:09:17,498 INFO    MainThread:1371664 [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': 150, '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': 'situglu', 'waleed_beta': 10.0, 'model_type': 'tiny_llama', 'output_attentions': False, 'output_dir': 'out/glu-situglu-150L_run', 'per_device_train_batch_size': 64, 'num_train_epochs': 1, 'max_steps': 1000, 'learning_rate': 0.005, 'lr_scheduler_type': 'constant_with_warmup', 'lr_scheduler_kwargs': None, 'warmup_steps': 200, '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': 0.0, '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-situglu-150L-25.1M-20260812-200915', 'project': 'huggingface', 'trackio_space_id': None, 'trackio_bucket_id': None, 'trackio_static_space_id': None, 'eval_strategy': 'steps', 'eval_steps': 50, 'eval_delay': 0, 'per_device_eval_batch_size': 128, 'prediction_loss_only': False, 'eval_on_start': False, 'eval_do_concat_batches': True, 'eval_use_gather_object': False, 'eval_accumulation_steps': None, 'include_for_metrics': [], 'batch_eval_metrics': False, 'save_only_model': False, 'save_strategy': 'steps', 'save_steps': 1000, '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-situglu-150L', '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}
-2026-08-12 20:09:17,506 INFO    MainThread:1371664 [wandb_config.py:__setitem__():155] [no run ID] config set model/num_parameters = 25138816 - <bound method Run._config_callback of <wandb.sdk.wandb_run.Run object at 0x14c63c1ee8d0>>
-2026-08-12 20:09:17,506 INFO    MainThread:1371664 [wandb_run.py:_config_callback():1346] config_cb model/num_parameters 25138816 None
+2026-08-12 20:24:59,715 INFO    MainThread:1371664 [wandb_init.py:init():820] starting backend
+2026-08-12 20:24:59,715 INFO    MainThread:1371664 [wandb_init.py:init():826] Connected to an existing wandb-core service via WANDB_SERVICE
+2026-08-12 20:24:59,715 INFO    MainThread:1371664 [wandb_init.py:init():835] sending inform_init request
+2026-08-12 20:24:59,979 INFO    MainThread:1371664 [wandb_init.py:init():840] backend started and connected
+2026-08-12 20:24:59,980 INFO    MainThread:1371664 [wandb_init.py:init():910] updated telemetry
+2026-08-12 20:24:59,987 INFO    MainThread:1371664 [wandb_init.py:init():933] communicating run to backend with 90.0 second timeout
+2026-08-12 20:25:00,444 INFO    MainThread:1371664 [wandb_init.py:init():978] starting run threads in backend
+2026-08-12 20:25:00,537 INFO    MainThread:1371664 [wandb_run.py:_console_start():2621] atexit reg
+2026-08-12 20:25:00,537 INFO    MainThread:1371664 [wandb_run.py:_redirect():2471] redirect: wrap_raw
+2026-08-12 20:25:00,537 INFO    MainThread:1371664 [wandb_run.py:_redirect():2540] Wrapping output streams.
+2026-08-12 20:25:00,537 INFO    MainThread:1371664 [wandb_run.py:_redirect():2563] Redirects installed.
+2026-08-12 20:25:00,538 INFO    MainThread:1371664 [wandb_init.py:init():1016] run started, returning control to user process
+2026-08-12 20:25:00,539 INFO    MainThread:1371664 [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': 150, 'num_attention_heads': 4, 'num_key_value_heads': 4, 'hidden_act': 'silu', 'max_position_embeddings': 512, 'initializer_range': 0.02, 'rms_norm_eps': 1e-06, 'use_cache': False, 'pad_token_id': 0, 'bos_token_id': 1, 'eos_token_id': 2, 'pretraining_tp': 1, 'tie_word_embeddings': True, 'rope_parameters': {'rope_theta': 10000.0, 'rope_type': 'default'}, 'attention_bias': False, 'attention_dropout': 0.0, 'mlp_bias': False, 'head_dim': 32, '_name_or_path': '', 'tokenizer_name': 'w-ahmad/tiny-stories-tokenizer', 'mlp_type': 'glu', 'activation': 'waleed', 'waleed_beta': 10.0, 'model_type': 'tiny_llama', 'output_attentions': False, 'output_dir': 'out/glu-waleed-150L_run', 'per_device_train_batch_size': 64, 'num_train_epochs': 1, 'max_steps': 1000, 'learning_rate': 0.005, 'lr_scheduler_type': 'constant_with_warmup', 'lr_scheduler_kwargs': None, 'warmup_steps': 200, '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': 0.0, 'label_smoothing_factor': 0.0, 'bf16': True, 'fp16': False, 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'gradient_checkpointing': False, 'gradient_checkpointing_kwargs': None, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'use_liger_kernel': False, 'liger_kernel_config': None, 'neftune_noise_alpha': None, 'torch_empty_cache_steps': None, 'auto_find_batch_size': False, 'logging_strategy': 'steps', 'logging_steps': 20, 'logging_first_step': False, 'log_on_each_node': True, 'logging_nan_inf_filter': True, 'include_num_input_tokens_seen': 'no', 'log_level': 'passive', 'log_level_replica': 'warning', 'disable_tqdm': False, 'report_to': ['wandb'], 'run_name': 'LM-glu-waleed-150L-25.1M-20260812-202458', 'project': 'huggingface', 'trackio_space_id': None, 'trackio_bucket_id': None, 'trackio_static_space_id': None, 'eval_strategy': 'steps', 'eval_steps': 50, 'eval_delay': 0, 'per_device_eval_batch_size': 128, 'prediction_loss_only': False, 'eval_on_start': False, 'eval_do_concat_batches': True, 'eval_use_gather_object': False, 'eval_accumulation_steps': None, 'include_for_metrics': [], 'batch_eval_metrics': False, 'save_only_model': False, 'save_strategy': 'steps', 'save_steps': 1000, '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-waleed-150L', '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}
+2026-08-12 20:25:00,546 INFO    MainThread:1371664 [wandb_config.py:__setitem__():155] [no run ID] config set model/num_parameters = 25138816 - <bound method Run._config_callback of <wandb.sdk.wandb_run.Run object at 0x14c663f22810>>
+2026-08-12 20:25:00,546 INFO    MainThread:1371664 [wandb_run.py:_config_callback():1346] config_cb model/num_parameters 25138816 None

zain/Activation/wandb/run-20260812_175304-hrggki9k/logs/debug-core.log

@@ -81,3 +81,10 @@
 {"time":"2026-08-12T20:09:16.738378339Z","level":"INFO","msg":"handleInformInit: received","streamId":"iwodu993","id":"3(@)"}
 {"time":"2026-08-12T20:09:16.998083611Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"iwodu993","id":"3(@)"}
 {"time":"2026-08-12T20:09:22.499682141Z","level":"INFO","msg":"connection: cancelling request","id":"3(@)","requestId":"jqt66vzz60j0"}
+{"time":"2026-08-12T20:24:57.908215673Z","level":"INFO","msg":"connection: cancelling request","id":"3(@)","requestId":"jqt66vzz60j0"}
+{"time":"2026-08-12T20:24:58.426858967Z","level":"INFO","msg":"connection: cancelling request","id":"3(@)","requestId":"jqt66vzz60j0"}
+{"time":"2026-08-12T20:24:58.428337581Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"iwodu993","id":"3(@)"}
+{"time":"2026-08-12T20:24:58.4291701Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"iwodu993","id":"3(@)"}
+{"time":"2026-08-12T20:24:59.716360765Z","level":"INFO","msg":"handleInformInit: received","streamId":"k14eff4o","id":"3(@)"}
+{"time":"2026-08-12T20:24:59.979099576Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"k14eff4o","id":"3(@)"}
+{"time":"2026-08-12T20:25:05.540298401Z","level":"INFO","msg":"connection: cancelling request","id":"3(@)","requestId":"dhf762wounvi"}

zain/Activation/wandb/run-20260812_180713-utqzlg67/logs/debug-core.log

@@ -81,3 +81,10 @@
 {"time":"2026-08-12T20:09:16.738378339Z","level":"INFO","msg":"handleInformInit: received","streamId":"iwodu993","id":"3(@)"}
 {"time":"2026-08-12T20:09:16.998083611Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"iwodu993","id":"3(@)"}
 {"time":"2026-08-12T20:09:22.499682141Z","level":"INFO","msg":"connection: cancelling request","id":"3(@)","requestId":"jqt66vzz60j0"}
+{"time":"2026-08-12T20:24:57.908215673Z","level":"INFO","msg":"connection: cancelling request","id":"3(@)","requestId":"jqt66vzz60j0"}
+{"time":"2026-08-12T20:24:58.426858967Z","level":"INFO","msg":"connection: cancelling request","id":"3(@)","requestId":"jqt66vzz60j0"}
+{"time":"2026-08-12T20:24:58.428337581Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"iwodu993","id":"3(@)"}
+{"time":"2026-08-12T20:24:58.4291701Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"iwodu993","id":"3(@)"}
+{"time":"2026-08-12T20:24:59.716360765Z","level":"INFO","msg":"handleInformInit: received","streamId":"k14eff4o","id":"3(@)"}
+{"time":"2026-08-12T20:24:59.979099576Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"k14eff4o","id":"3(@)"}
+{"time":"2026-08-12T20:25:05.540298401Z","level":"INFO","msg":"connection: cancelling request","id":"3(@)","requestId":"dhf762wounvi"}

zain/Activation/wandb/run-20260812_182126-vyycsg6r/logs/debug-core.log

@@ -81,3 +81,10 @@
 {"time":"2026-08-12T20:09:16.738378339Z","level":"INFO","msg":"handleInformInit: received","streamId":"iwodu993","id":"3(@)"}
 {"time":"2026-08-12T20:09:16.998083611Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"iwodu993","id":"3(@)"}
 {"time":"2026-08-12T20:09:22.499682141Z","level":"INFO","msg":"connection: cancelling request","id":"3(@)","requestId":"jqt66vzz60j0"}
+{"time":"2026-08-12T20:24:57.908215673Z","level":"INFO","msg":"connection: cancelling request","id":"3(@)","requestId":"jqt66vzz60j0"}
+{"time":"2026-08-12T20:24:58.426858967Z","level":"INFO","msg":"connection: cancelling request","id":"3(@)","requestId":"jqt66vzz60j0"}
+{"time":"2026-08-12T20:24:58.428337581Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"iwodu993","id":"3(@)"}
+{"time":"2026-08-12T20:24:58.4291701Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"iwodu993","id":"3(@)"}
+{"time":"2026-08-12T20:24:59.716360765Z","level":"INFO","msg":"handleInformInit: received","streamId":"k14eff4o","id":"3(@)"}
+{"time":"2026-08-12T20:24:59.979099576Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"k14eff4o","id":"3(@)"}
+{"time":"2026-08-12T20:25:05.540298401Z","level":"INFO","msg":"connection: cancelling request","id":"3(@)","requestId":"dhf762wounvi"}

zain/Activation/wandb/run-20260812_183823-qtstmmvv/logs/debug-core.log

@@ -81,3 +81,10 @@
 {"time":"2026-08-12T20:09:16.738378339Z","level":"INFO","msg":"handleInformInit: received","streamId":"iwodu993","id":"3(@)"}
 {"time":"2026-08-12T20:09:16.998083611Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"iwodu993","id":"3(@)"}
 {"time":"2026-08-12T20:09:22.499682141Z","level":"INFO","msg":"connection: cancelling request","id":"3(@)","requestId":"jqt66vzz60j0"}
+{"time":"2026-08-12T20:24:57.908215673Z","level":"INFO","msg":"connection: cancelling request","id":"3(@)","requestId":"jqt66vzz60j0"}
+{"time":"2026-08-12T20:24:58.426858967Z","level":"INFO","msg":"connection: cancelling request","id":"3(@)","requestId":"jqt66vzz60j0"}
+{"time":"2026-08-12T20:24:58.428337581Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"iwodu993","id":"3(@)"}
+{"time":"2026-08-12T20:24:58.4291701Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"iwodu993","id":"3(@)"}
+{"time":"2026-08-12T20:24:59.716360765Z","level":"INFO","msg":"handleInformInit: received","streamId":"k14eff4o","id":"3(@)"}
+{"time":"2026-08-12T20:24:59.979099576Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"k14eff4o","id":"3(@)"}
+{"time":"2026-08-12T20:25:05.540298401Z","level":"INFO","msg":"connection: cancelling request","id":"3(@)","requestId":"dhf762wounvi"}