| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.999980000399992, | |
| "global_step": 50000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 1.2e-06, | |
| "loss": 16.0574, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 0.0003, | |
| "loss": 3.0097, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.0006, | |
| "loss": 1.2654, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.0005999622383021625, | |
| "loss": 1.1616, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.0005998489627149555, | |
| "loss": 1.1163, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.0005996602017549024, | |
| "loss": 1.088, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.0005993960029415653, | |
| "loss": 1.0675, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.0005990564327855827, | |
| "loss": 1.0509, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.0005986415767719254, | |
| "loss": 1.0367, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.0005981515393383762, | |
| "loss": 1.0255, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.0005975864438492385, | |
| "loss": 1.0155, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "eval_loss": 0.9499261975288391, | |
| "eval_runtime": 327.1182, | |
| "eval_samples_per_second": 313.037, | |
| "eval_steps_per_second": 4.891, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.0005969464325642798, | |
| "loss": 1.0062, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.0005962316666029183, | |
| "loss": 0.9966, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.0005954423259036624, | |
| "loss": 0.9894, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0005945786091788119, | |
| "loss": 0.9824, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0005936407338644336, | |
| "loss": 0.9762, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.0005926289360656221, | |
| "loss": 0.9716, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.0005915434704970625, | |
| "loss": 0.9658, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.0005903846104189068, | |
| "loss": 0.9599, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.0005891526475679825, | |
| "loss": 0.9559, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.0005878478920843492, | |
| "loss": 0.9512, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_loss": 0.9004160165786743, | |
| "eval_runtime": 327.5819, | |
| "eval_samples_per_second": 312.594, | |
| "eval_steps_per_second": 4.884, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.0005864706724332221, | |
| "loss": 0.9472, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.0005850213353222835, | |
| "loss": 0.9432, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.0005835002456144005, | |
| "loss": 0.9404, | |
| "step": 5750 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.0005819077862357724, | |
| "loss": 0.9364, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.000580244358079532, | |
| "loss": 0.9336, | |
| "step": 6250 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.0005785103799048218, | |
| "loss": 0.93, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.0005767062882313744, | |
| "loss": 0.9265, | |
| "step": 6750 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.0005748325372296208, | |
| "loss": 0.9245, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.0005728895986063555, | |
| "loss": 0.9221, | |
| "step": 7250 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0005708779614859863, | |
| "loss": 0.9185, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 0.8724892139434814, | |
| "eval_runtime": 327.0613, | |
| "eval_samples_per_second": 313.091, | |
| "eval_steps_per_second": 4.892, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0005687981322874007, | |
| "loss": 0.9166, | |
| "step": 7750 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.000566650634596477, | |
| "loss": 0.9147, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.0005644360090342746, | |
| "loss": 0.9121, | |
| "step": 8250 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.0005621548131209354, | |
| "loss": 0.9103, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.0005598076211353316, | |
| "loss": 0.9078, | |
| "step": 8750 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.000557395023970493, | |
| "loss": 0.9056, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.0005549176289848543, | |
| "loss": 0.904, | |
| "step": 9250 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.0005523760598493544, | |
| "loss": 0.9027, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.0005497709563904314, | |
| "loss": 0.9009, | |
| "step": 9750 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0005471029744289498, | |
| "loss": 0.8991, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_loss": 0.8550655841827393, | |
| "eval_runtime": 327.3543, | |
| "eval_samples_per_second": 312.811, | |
| "eval_steps_per_second": 4.888, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0005443727856151006, | |
| "loss": 0.8972, | |
| "step": 10250 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.0005415810772593175, | |
| "loss": 0.8956, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.0005387285521592496, | |
| "loss": 0.8941, | |
| "step": 10750 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0005358159284228363, | |
| "loss": 0.8926, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.000532843939287527, | |
| "loss": 0.8907, | |
| "step": 11250 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.0005298133329356933, | |
| "loss": 0.889, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.0005267248723062775, | |
| "loss": 0.888, | |
| "step": 11750 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.0005235793349027264, | |
| "loss": 0.8858, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.0005203775125972599, | |
| "loss": 0.8851, | |
| "step": 12250 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.000517120211431521, | |
| "loss": 0.8838, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 0.8417796492576599, | |
| "eval_runtime": 327.2294, | |
| "eval_samples_per_second": 312.93, | |
| "eval_steps_per_second": 4.89, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.0005138082514136589, | |
| "loss": 0.8821, | |
| "step": 12750 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.0005104424663118964, | |
| "loss": 0.8814, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.0005070237034446336, | |
| "loss": 0.8801, | |
| "step": 13250 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.0005035528234671396, | |
| "loss": 0.8794, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.0005000307001548875, | |
| "loss": 0.8782, | |
| "step": 13750 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.0004964582201835855, | |
| "loss": 0.8765, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.0004928362829059618, | |
| "loss": 0.8756, | |
| "step": 14250 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0004891658001253567, | |
| "loss": 0.8741, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.00048544769586618153, | |
| "loss": 0.8728, | |
| "step": 14750 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.00048168290614129995, | |
| "loss": 0.8729, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_loss": 0.8313615322113037, | |
| "eval_runtime": 327.2078, | |
| "eval_samples_per_second": 312.951, | |
| "eval_steps_per_second": 4.89, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.00047787237871639213, | |
| "loss": 0.8719, | |
| "step": 15250 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0004740170728713594, | |
| "loss": 0.8703, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0004701179591588311, | |
| "loss": 0.8701, | |
| "step": 15750 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.00046617601915983307, | |
| "loss": 0.8684, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.00046219224523667927, | |
| "loss": 0.8684, | |
| "step": 16250 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.00045816764028315066, | |
| "loss": 0.8676, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.0004541032174720219, | |
| "loss": 0.8663, | |
| "step": 16750 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.00045, | |
| "loss": 0.8653, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.00044585902083014057, | |
| "loss": 0.8644, | |
| "step": 17250 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.0004416813224318048, | |
| "loss": 0.8634, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "eval_loss": 0.8233901858329773, | |
| "eval_runtime": 326.8294, | |
| "eval_samples_per_second": 313.313, | |
| "eval_steps_per_second": 4.896, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.00043746795651822306, | |
| "loss": 0.863, | |
| "step": 17750 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0004332199837817322, | |
| "loss": 0.8621, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0004289384736267515, | |
| "loss": 0.8616, | |
| "step": 18250 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.00042462450390056593, | |
| "loss": 0.861, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.0004202791606219841, | |
| "loss": 0.8597, | |
| "step": 18750 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.0004159035377079385, | |
| "loss": 0.8593, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.0004114987366980982, | |
| "loss": 0.8588, | |
| "step": 19250 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.0004070658664775615, | |
| "loss": 0.8586, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.00040260604299770063, | |
| "loss": 0.8569, | |
| "step": 19750 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0003981203889952265, | |
| "loss": 0.8565, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 0.817196786403656, | |
| "eval_runtime": 326.9673, | |
| "eval_samples_per_second": 313.181, | |
| "eval_steps_per_second": 4.893, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0003936100337095461, | |
| "loss": 0.8557, | |
| "step": 20250 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.0003890761125984825, | |
| "loss": 0.8547, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.0003845197670524289, | |
| "loss": 0.8546, | |
| "step": 20750 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.0003799421441070104, | |
| "loss": 0.8537, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.0003753443961543237, | |
| "loss": 0.8539, | |
| "step": 21250 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.0003707276806528282, | |
| "loss": 0.8528, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.0003660931598359622, | |
| "loss": 0.8521, | |
| "step": 21750 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0003614420004195572, | |
| "loss": 0.8518, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.000356775373308123, | |
| "loss": 0.8516, | |
| "step": 22250 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.0003520944533000791, | |
| "loss": 0.851, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "eval_loss": 0.8120535016059875, | |
| "eval_runtime": 326.9259, | |
| "eval_samples_per_second": 313.221, | |
| "eval_steps_per_second": 4.894, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.00034740041879200497, | |
| "loss": 0.8491, | |
| "step": 22750 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.00034269445148198553, | |
| "loss": 0.8492, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.00033797773607212474, | |
| "loss": 0.8483, | |
| "step": 23250 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.0003332514599703033, | |
| "loss": 0.8483, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.0003285168129912547, | |
| "loss": 0.8472, | |
| "step": 23750 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.0003237749870570365, | |
| "loss": 0.8462, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.0003190271758969693, | |
| "loss": 0.846, | |
| "step": 24250 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.00031427457474712274, | |
| "loss": 0.8471, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.0003095183800494203, | |
| "loss": 0.8458, | |
| "step": 24750 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.00030475978915044235, | |
| "loss": 0.8445, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 0.8070387244224548, | |
| "eval_runtime": 327.529, | |
| "eval_samples_per_second": 312.644, | |
| "eval_steps_per_second": 4.885, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.0003, | |
| "loss": 0.845, | |
| "step": 25250 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.0002952402108495576, | |
| "loss": 0.8442, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.00029048161995057974, | |
| "loss": 0.8445, | |
| "step": 25750 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.0002857254252528773, | |
| "loss": 0.8436, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.00028097282410303066, | |
| "loss": 0.8434, | |
| "step": 26250 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.0002762250129429634, | |
| "loss": 0.8419, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.00027148318700874523, | |
| "loss": 0.8419, | |
| "step": 26750 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.0002667485400296967, | |
| "loss": 0.8423, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.00026202226392787515, | |
| "loss": 0.8414, | |
| "step": 27250 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 0.0002573055485180145, | |
| "loss": 0.8408, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "eval_loss": 0.8027185797691345, | |
| "eval_runtime": 327.1182, | |
| "eval_samples_per_second": 313.037, | |
| "eval_steps_per_second": 4.891, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 0.000252599581207995, | |
| "loss": 0.8412, | |
| "step": 27750 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.0002479055466999209, | |
| "loss": 0.8396, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.00024322462669187702, | |
| "loss": 0.8396, | |
| "step": 28250 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.0002385579995804428, | |
| "loss": 0.8391, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.00023390684016403777, | |
| "loss": 0.8389, | |
| "step": 28750 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.00022927231934717176, | |
| "loss": 0.8391, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.00022465560384567624, | |
| "loss": 0.8384, | |
| "step": 29250 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.00022005785589298952, | |
| "loss": 0.8374, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.00021548023294757105, | |
| "loss": 0.8378, | |
| "step": 29750 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.00021092388740151762, | |
| "loss": 0.8366, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_loss": 0.7997037768363953, | |
| "eval_runtime": 327.6706, | |
| "eval_samples_per_second": 312.509, | |
| "eval_steps_per_second": 4.883, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.00020638996629045387, | |
| "loss": 0.8368, | |
| "step": 30250 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 0.0002018796110047735, | |
| "loss": 0.837, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 0.00019739395700229937, | |
| "loss": 0.8362, | |
| "step": 30750 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.00019293413352243846, | |
| "loss": 0.8363, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.00018850126330190176, | |
| "loss": 0.8364, | |
| "step": 31250 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.00018409646229206137, | |
| "loss": 0.835, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.00017972083937801593, | |
| "loss": 0.8349, | |
| "step": 31750 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 0.0001753754960994341, | |
| "loss": 0.8349, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 0.0001710615263732485, | |
| "loss": 0.8347, | |
| "step": 32250 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.00016678001621826772, | |
| "loss": 0.8336, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "eval_loss": 0.7965421080589294, | |
| "eval_runtime": 326.8259, | |
| "eval_samples_per_second": 313.317, | |
| "eval_steps_per_second": 4.896, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.00016253204348177686, | |
| "loss": 0.8338, | |
| "step": 32750 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.00015831867756819522, | |
| "loss": 0.834, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.00015414097916985944, | |
| "loss": 0.8338, | |
| "step": 33250 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.00015000000000000004, | |
| "loss": 0.8321, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.00014589678252797817, | |
| "loss": 0.8328, | |
| "step": 33750 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.00014183235971684924, | |
| "loss": 0.8323, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.00013780775476332082, | |
| "loss": 0.8318, | |
| "step": 34250 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.0001338239808401669, | |
| "loss": 0.832, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.0001298820408411688, | |
| "loss": 0.8319, | |
| "step": 34750 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.00012598292712864058, | |
| "loss": 0.8308, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "eval_loss": 0.7937933206558228, | |
| "eval_runtime": 326.8361, | |
| "eval_samples_per_second": 313.307, | |
| "eval_steps_per_second": 4.895, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.0001221276212836079, | |
| "loss": 0.8315, | |
| "step": 35250 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.00011831709385870004, | |
| "loss": 0.831, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.0001145523041338184, | |
| "loss": 0.8307, | |
| "step": 35750 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.00011083419987464334, | |
| "loss": 0.83, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.00010716371709403818, | |
| "loss": 0.8309, | |
| "step": 36250 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.00010354177981641449, | |
| "loss": 0.8305, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 9.996929984511254e-05, | |
| "loss": 0.8299, | |
| "step": 36750 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 9.644717653286037e-05, | |
| "loss": 0.829, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 9.297629655536644e-05, | |
| "loss": 0.8296, | |
| "step": 37250 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 8.955753368810358e-05, | |
| "loss": 0.8292, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eval_loss": 0.7918949127197266, | |
| "eval_runtime": 327.7211, | |
| "eval_samples_per_second": 312.461, | |
| "eval_steps_per_second": 4.882, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 8.619174858634122e-05, | |
| "loss": 0.8293, | |
| "step": 37750 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 8.287978856847894e-05, | |
| "loss": 0.8297, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 7.962248740274003e-05, | |
| "loss": 0.8283, | |
| "step": 38250 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 7.642066509727359e-05, | |
| "loss": 0.8283, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 7.327512769372254e-05, | |
| "loss": 0.8286, | |
| "step": 38750 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 7.018666706430662e-05, | |
| "loss": 0.8281, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 6.715606071247291e-05, | |
| "loss": 0.8282, | |
| "step": 39250 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 6.418407157716381e-05, | |
| "loss": 0.8285, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 6.127144784075033e-05, | |
| "loss": 0.828, | |
| "step": 39750 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 5.841892274068241e-05, | |
| "loss": 0.8272, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 0.7908062934875488, | |
| "eval_runtime": 327.1979, | |
| "eval_samples_per_second": 312.96, | |
| "eval_steps_per_second": 4.89, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 5.562721438489928e-05, | |
| "loss": 0.8279, | |
| "step": 40250 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 5.2897025571050186e-05, | |
| "loss": 0.8273, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 5.022904360956861e-05, | |
| "loss": 0.8276, | |
| "step": 40750 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 4.76239401506456e-05, | |
| "loss": 0.8275, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 4.5082371015145716e-05, | |
| "loss": 0.8271, | |
| "step": 41250 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 4.260497602950688e-05, | |
| "loss": 0.8266, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 4.019237886466838e-05, | |
| "loss": 0.827, | |
| "step": 41750 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 3.784518687906452e-05, | |
| "loss": 0.8268, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 3.556399096572541e-05, | |
| "loss": 0.8267, | |
| "step": 42250 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 3.3349365403522986e-05, | |
| "loss": 0.8268, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "eval_loss": 0.7894444465637207, | |
| "eval_runtime": 327.0054, | |
| "eval_samples_per_second": 313.145, | |
| "eval_steps_per_second": 4.893, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 3.120186771259927e-05, | |
| "loss": 0.8266, | |
| "step": 42750 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 2.9122038514013678e-05, | |
| "loss": 0.8264, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 2.7110401393644464e-05, | |
| "loss": 0.8265, | |
| "step": 43250 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 2.516746277037912e-05, | |
| "loss": 0.8259, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 2.329371176862562e-05, | |
| "loss": 0.8258, | |
| "step": 43750 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 2.148962009517823e-05, | |
| "loss": 0.8257, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 1.9755641920468003e-05, | |
| "loss": 0.8266, | |
| "step": 44250 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 1.8092213764227503e-05, | |
| "loss": 0.8266, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 1.6499754385599462e-05, | |
| "loss": 0.826, | |
| "step": 44750 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 1.4978664677716402e-05, | |
| "loss": 0.8254, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "eval_loss": 0.7888011336326599, | |
| "eval_runtime": 326.8978, | |
| "eval_samples_per_second": 313.248, | |
| "eval_steps_per_second": 4.894, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 1.3529327566777836e-05, | |
| "loss": 0.8256, | |
| "step": 45250 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 1.2152107915650821e-05, | |
| "loss": 0.8265, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 1.0847352432017387e-05, | |
| "loss": 0.8254, | |
| "step": 45750 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 9.615389581093124e-06, | |
| "loss": 0.8251, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 8.456529502937504e-06, | |
| "loss": 0.8253, | |
| "step": 46250 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 7.371063934377885e-06, | |
| "loss": 0.8258, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 6.35926613556641e-06, | |
| "loss": 0.8251, | |
| "step": 46750 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 5.421390821187988e-06, | |
| "loss": 0.825, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 4.557674096337593e-06, | |
| "loss": 0.8247, | |
| "step": 47250 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 3.768333397081713e-06, | |
| "loss": 0.8259, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "eval_loss": 0.7886191010475159, | |
| "eval_runtime": 326.893, | |
| "eval_samples_per_second": 313.252, | |
| "eval_steps_per_second": 4.895, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 3.0535674357201944e-06, | |
| "loss": 0.826, | |
| "step": 47750 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 2.4135561507613975e-06, | |
| "loss": 0.8251, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 1.848460661623763e-06, | |
| "loss": 0.8253, | |
| "step": 48250 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 1.3584232280746231e-06, | |
| "loss": 0.8252, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 9.435672144173178e-07, | |
| "loss": 0.8249, | |
| "step": 48750 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 6.03997058434702e-07, | |
| "loss": 0.8243, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 3.397982450976111e-07, | |
| "loss": 0.8249, | |
| "step": 49250 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 1.5103728504447522e-07, | |
| "loss": 0.8251, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 3.776169783747951e-08, | |
| "loss": 0.8247, | |
| "step": 49750 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.825, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.7884970307350159, | |
| "eval_runtime": 326.7465, | |
| "eval_samples_per_second": 313.393, | |
| "eval_steps_per_second": 4.897, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 50000, | |
| "total_flos": 1.147093508736575e+19, | |
| "train_loss": 0.8805710920715332, | |
| "train_runtime": 251252.3922, | |
| "train_samples_per_second": 101.89, | |
| "train_steps_per_second": 0.199 | |
| } | |
| ], | |
| "max_steps": 50000, | |
| "num_train_epochs": 1, | |
| "total_flos": 1.147093508736575e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |