| { | |
| "best_metric": 1.1513956785202026, | |
| "best_model_checkpoint": "data/Mistral-7B_task-1_120-samples_config-1/checkpoint-33", | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 110, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09090909090909091, | |
| "grad_norm": 6.151435375213623, | |
| "learning_rate": 1.818181818181818e-06, | |
| "loss": 2.4652, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.18181818181818182, | |
| "grad_norm": 6.202335834503174, | |
| "learning_rate": 3.636363636363636e-06, | |
| "loss": 2.5638, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.36363636363636365, | |
| "grad_norm": 5.723293304443359, | |
| "learning_rate": 7.272727272727272e-06, | |
| "loss": 2.2237, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.5454545454545454, | |
| "grad_norm": 5.8827128410339355, | |
| "learning_rate": 1.0909090909090909e-05, | |
| "loss": 2.1827, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 5.391604900360107, | |
| "learning_rate": 1.4545454545454545e-05, | |
| "loss": 2.2469, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.9090909090909091, | |
| "grad_norm": 4.755679130554199, | |
| "learning_rate": 1.8181818181818182e-05, | |
| "loss": 1.8479, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.7766928672790527, | |
| "eval_runtime": 9.6489, | |
| "eval_samples_per_second": 2.487, | |
| "eval_steps_per_second": 2.487, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.0909090909090908, | |
| "grad_norm": 3.921355724334717, | |
| "learning_rate": 2.1818181818181818e-05, | |
| "loss": 1.7206, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.2727272727272727, | |
| "grad_norm": 3.4946775436401367, | |
| "learning_rate": 2.5454545454545454e-05, | |
| "loss": 1.7337, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.4545454545454546, | |
| "grad_norm": 2.374727487564087, | |
| "learning_rate": 2.909090909090909e-05, | |
| "loss": 1.4858, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6363636363636362, | |
| "grad_norm": 2.04058575630188, | |
| "learning_rate": 3.272727272727273e-05, | |
| "loss": 1.3494, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8181818181818183, | |
| "grad_norm": 2.1711928844451904, | |
| "learning_rate": 3.6363636363636364e-05, | |
| "loss": 1.3026, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.4772785902023315, | |
| "learning_rate": 4e-05, | |
| "loss": 1.3159, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.3060346841812134, | |
| "eval_runtime": 9.6501, | |
| "eval_samples_per_second": 2.487, | |
| "eval_steps_per_second": 2.487, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.1818181818181817, | |
| "grad_norm": 1.4939863681793213, | |
| "learning_rate": 4.3636363636363636e-05, | |
| "loss": 1.2751, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.3636363636363638, | |
| "grad_norm": 1.551141381263733, | |
| "learning_rate": 4.7272727272727275e-05, | |
| "loss": 1.251, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.5454545454545454, | |
| "grad_norm": 1.5137313604354858, | |
| "learning_rate": 5.090909090909091e-05, | |
| "loss": 1.0985, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.7272727272727275, | |
| "grad_norm": 1.3787038326263428, | |
| "learning_rate": 5.4545454545454546e-05, | |
| "loss": 1.0334, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.909090909090909, | |
| "grad_norm": 1.6689972877502441, | |
| "learning_rate": 5.818181818181818e-05, | |
| "loss": 1.0602, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 1.1513956785202026, | |
| "eval_runtime": 9.6608, | |
| "eval_samples_per_second": 2.484, | |
| "eval_steps_per_second": 2.484, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.090909090909091, | |
| "grad_norm": 1.6474758386611938, | |
| "learning_rate": 6.181818181818182e-05, | |
| "loss": 1.0549, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.2727272727272725, | |
| "grad_norm": 1.4157817363739014, | |
| "learning_rate": 6.545454545454546e-05, | |
| "loss": 0.9351, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.4545454545454546, | |
| "grad_norm": 1.760252833366394, | |
| "learning_rate": 6.90909090909091e-05, | |
| "loss": 0.9018, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.6363636363636362, | |
| "grad_norm": 1.7269090414047241, | |
| "learning_rate": 7.272727272727273e-05, | |
| "loss": 0.8603, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.8181818181818183, | |
| "grad_norm": 1.7982224225997925, | |
| "learning_rate": 7.636363636363637e-05, | |
| "loss": 0.8153, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 1.8845890760421753, | |
| "learning_rate": 8e-05, | |
| "loss": 0.7949, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.172354817390442, | |
| "eval_runtime": 9.6525, | |
| "eval_samples_per_second": 2.486, | |
| "eval_steps_per_second": 2.486, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.181818181818182, | |
| "grad_norm": 1.8724236488342285, | |
| "learning_rate": 8.363636363636364e-05, | |
| "loss": 0.6183, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.363636363636363, | |
| "grad_norm": 2.1722185611724854, | |
| "learning_rate": 8.727272727272727e-05, | |
| "loss": 0.5739, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.545454545454545, | |
| "grad_norm": 2.6997480392456055, | |
| "learning_rate": 9.090909090909092e-05, | |
| "loss": 0.5654, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.7272727272727275, | |
| "grad_norm": 2.968306541442871, | |
| "learning_rate": 9.454545454545455e-05, | |
| "loss": 0.5868, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 4.909090909090909, | |
| "grad_norm": 2.91367244720459, | |
| "learning_rate": 9.818181818181818e-05, | |
| "loss": 0.5137, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 1.346767783164978, | |
| "eval_runtime": 9.6556, | |
| "eval_samples_per_second": 2.486, | |
| "eval_steps_per_second": 2.486, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.090909090909091, | |
| "grad_norm": 2.089282274246216, | |
| "learning_rate": 9.999899300364532e-05, | |
| "loss": 0.421, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.2727272727272725, | |
| "grad_norm": 2.366018772125244, | |
| "learning_rate": 9.99909372761763e-05, | |
| "loss": 0.274, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.454545454545454, | |
| "grad_norm": 2.2366552352905273, | |
| "learning_rate": 9.997482711915927e-05, | |
| "loss": 0.27, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.636363636363637, | |
| "grad_norm": 2.655911684036255, | |
| "learning_rate": 9.99506651282272e-05, | |
| "loss": 0.2622, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 5.818181818181818, | |
| "grad_norm": 4.180952548980713, | |
| "learning_rate": 9.991845519630678e-05, | |
| "loss": 0.2599, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 3.6061222553253174, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.2333, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 1.6914259195327759, | |
| "eval_runtime": 9.6466, | |
| "eval_samples_per_second": 2.488, | |
| "eval_steps_per_second": 2.488, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.181818181818182, | |
| "grad_norm": 2.1230361461639404, | |
| "learning_rate": 9.982991356370404e-05, | |
| "loss": 0.1391, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.363636363636363, | |
| "grad_norm": 3.146448850631714, | |
| "learning_rate": 9.977359612865423e-05, | |
| "loss": 0.1079, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.545454545454545, | |
| "grad_norm": 3.548719644546509, | |
| "learning_rate": 9.970925928158274e-05, | |
| "loss": 0.1151, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 6.7272727272727275, | |
| "grad_norm": 2.8574233055114746, | |
| "learning_rate": 9.963691338830044e-05, | |
| "loss": 0.1403, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 6.909090909090909, | |
| "grad_norm": 3.1925230026245117, | |
| "learning_rate": 9.955657010501806e-05, | |
| "loss": 0.1177, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 1.8391222953796387, | |
| "eval_runtime": 9.6552, | |
| "eval_samples_per_second": 2.486, | |
| "eval_steps_per_second": 2.486, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 7.090909090909091, | |
| "grad_norm": 2.0694875717163086, | |
| "learning_rate": 9.946824237646824e-05, | |
| "loss": 0.0844, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 7.2727272727272725, | |
| "grad_norm": 2.7770867347717285, | |
| "learning_rate": 9.937194443381972e-05, | |
| "loss": 0.0774, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 7.454545454545454, | |
| "grad_norm": 1.8506356477737427, | |
| "learning_rate": 9.926769179238466e-05, | |
| "loss": 0.0646, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 7.636363636363637, | |
| "grad_norm": 2.3127522468566895, | |
| "learning_rate": 9.915550124911866e-05, | |
| "loss": 0.0926, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 7.818181818181818, | |
| "grad_norm": 1.9697074890136719, | |
| "learning_rate": 9.903539087991462e-05, | |
| "loss": 0.0568, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 1.8619533777236938, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.0738, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 1.8831233978271484, | |
| "eval_runtime": 9.6565, | |
| "eval_samples_per_second": 2.485, | |
| "eval_steps_per_second": 2.485, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 8.181818181818182, | |
| "grad_norm": 1.0307751893997192, | |
| "learning_rate": 9.877148934427037e-05, | |
| "loss": 0.0405, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 8.363636363636363, | |
| "grad_norm": 1.3500251770019531, | |
| "learning_rate": 9.862774069706346e-05, | |
| "loss": 0.0493, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 8.545454545454545, | |
| "grad_norm": 2.4816551208496094, | |
| "learning_rate": 9.847615725553456e-05, | |
| "loss": 0.0387, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 8.727272727272727, | |
| "grad_norm": 2.9809985160827637, | |
| "learning_rate": 9.831676344247342e-05, | |
| "loss": 0.0429, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 8.909090909090908, | |
| "grad_norm": 3.9411723613739014, | |
| "learning_rate": 9.814958493905963e-05, | |
| "loss": 0.0676, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 1.9783605337142944, | |
| "eval_runtime": 9.6493, | |
| "eval_samples_per_second": 2.487, | |
| "eval_steps_per_second": 2.487, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 9.090909090909092, | |
| "grad_norm": 1.2140812873840332, | |
| "learning_rate": 9.797464868072488e-05, | |
| "loss": 0.0389, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 9.272727272727273, | |
| "grad_norm": 0.9542932510375977, | |
| "learning_rate": 9.779198285281325e-05, | |
| "loss": 0.0228, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 9.454545454545455, | |
| "grad_norm": 3.229567289352417, | |
| "learning_rate": 9.760161688604008e-05, | |
| "loss": 0.0349, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 9.636363636363637, | |
| "grad_norm": 1.3410265445709229, | |
| "learning_rate": 9.740358145174998e-05, | |
| "loss": 0.023, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 9.818181818181818, | |
| "grad_norm": 2.039607048034668, | |
| "learning_rate": 9.719790845697533e-05, | |
| "loss": 0.0478, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 4.932439804077148, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.0588, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 2.13114857673645, | |
| "eval_runtime": 9.6508, | |
| "eval_samples_per_second": 2.487, | |
| "eval_steps_per_second": 2.487, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 110, | |
| "total_flos": 1.3256024760778752e+16, | |
| "train_loss": 0.674870581145991, | |
| "train_runtime": 1036.2269, | |
| "train_samples_per_second": 4.246, | |
| "train_steps_per_second": 0.531 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 550, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 7, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.3256024760778752e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |