| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9934065934065934, | |
| "eval_steps": 29, | |
| "global_step": 113, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 2e-05, | |
| "loss": 0.5065, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "eval_loss": 0.5089401602745056, | |
| "eval_runtime": 34.8867, | |
| "eval_samples_per_second": 174.68, | |
| "eval_steps_per_second": 12.498, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 4e-05, | |
| "loss": 0.5189, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 6e-05, | |
| "loss": 0.4966, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 8e-05, | |
| "loss": 0.4891, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.0001, | |
| "loss": 0.507, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.00012, | |
| "loss": 0.5071, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.00014, | |
| "loss": 0.4739, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.00016, | |
| "loss": 0.4721, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.00018, | |
| "loss": 0.4213, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.0002, | |
| "loss": 0.4056, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.00019999747405795057, | |
| "loss": 0.3989, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.00019998989635940995, | |
| "loss": 0.3843, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.00019997726728719468, | |
| "loss": 0.3771, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.00019995958747931082, | |
| "loss": 0.3775, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.00019993685782892183, | |
| "loss": 0.3778, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.00019990907948430327, | |
| "loss": 0.3786, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0001998762538487849, | |
| "loss": 0.3669, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.00019983838258067987, | |
| "loss": 0.3819, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.0001997954675932006, | |
| "loss": 0.4043, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.00019974751105436266, | |
| "loss": 0.3503, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.00019969451538687474, | |
| "loss": 0.3753, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.00019963648326801653, | |
| "loss": 0.368, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.00019957341762950344, | |
| "loss": 0.3828, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.00019950532165733845, | |
| "loss": 0.3509, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.00019943219879165115, | |
| "loss": 0.381, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.0001993540527265239, | |
| "loss": 0.3638, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.0001992708874098054, | |
| "loss": 0.3645, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.00019918270704291104, | |
| "loss": 0.3466, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.00019908951608061076, | |
| "loss": 0.3477, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 0.35608404874801636, | |
| "eval_runtime": 35.0585, | |
| "eval_samples_per_second": 173.824, | |
| "eval_steps_per_second": 12.436, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.000198991319230804, | |
| "loss": 0.3444, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.0001988881214542817, | |
| "loss": 0.3646, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.00019877992796447601, | |
| "loss": 0.3502, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.00019866674422719665, | |
| "loss": 0.3532, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.00019854857596035476, | |
| "loss": 0.3639, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0001984254291336743, | |
| "loss": 0.3538, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.0001982973099683902, | |
| "loss": 0.3431, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.00019816422493693417, | |
| "loss": 0.3493, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.0001980261807626078, | |
| "loss": 0.3596, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.00019788318441924275, | |
| "loss": 0.3401, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.00019773524313084856, | |
| "loss": 0.3822, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.00019758236437124769, | |
| "loss": 0.3719, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.00019742455586369784, | |
| "loss": 0.3369, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.0001972618255805019, | |
| "loss": 0.341, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.0001970941817426052, | |
| "loss": 0.3481, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.00019692163281918014, | |
| "loss": 0.3615, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.00019674418752719833, | |
| "loss": 0.346, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.00019656185483099024, | |
| "loss": 0.3479, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.00019637464394179239, | |
| "loss": 0.3638, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.00019618256431728194, | |
| "loss": 0.3419, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0001959856256610988, | |
| "loss": 0.3496, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.00019578383792235573, | |
| "loss": 0.3503, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.0001955772112951354, | |
| "loss": 0.3437, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.00019536575621797547, | |
| "loss": 0.3343, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.00019514948337334142, | |
| "loss": 0.3517, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.0001949284036870867, | |
| "loss": 0.343, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.0001947025283279008, | |
| "loss": 0.3479, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.00019447186870674506, | |
| "loss": 0.3237, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.00019423643647627624, | |
| "loss": 0.3593, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_loss": 0.3460778594017029, | |
| "eval_runtime": 35.0517, | |
| "eval_samples_per_second": 173.858, | |
| "eval_steps_per_second": 12.439, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.00019399624353025772, | |
| "loss": 0.348, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.00019375130200295878, | |
| "loss": 0.344, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.0001935016242685415, | |
| "loss": 0.3601, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 0.00019324722294043558, | |
| "loss": 0.3524, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 0.00019298811087070132, | |
| "loss": 0.3421, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.00019272430114938017, | |
| "loss": 0.3524, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.00019245580710383343, | |
| "loss": 0.3334, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.00019218264229806917, | |
| "loss": 0.3312, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.00019190482053205673, | |
| "loss": 0.344, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.00019162235584102973, | |
| "loss": 0.353, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 0.00019133526249477702, | |
| "loss": 0.3473, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.00019104355499692166, | |
| "loss": 0.345, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.00019074724808418835, | |
| "loss": 0.348, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.00019044635672565895, | |
| "loss": 0.3333, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 0.00019014089612201613, | |
| "loss": 0.3398, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.00018983088170477553, | |
| "loss": 0.3424, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.00018951632913550626, | |
| "loss": 0.3279, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.00018919725430503945, | |
| "loss": 0.3243, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.00018887367333266582, | |
| "loss": 0.3439, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.000188545602565321, | |
| "loss": 0.3486, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.00018821305857675996, | |
| "loss": 0.3367, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.00018787605816671953, | |
| "loss": 0.3386, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.00018753461836006988, | |
| "loss": 0.3574, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.00018718875640595434, | |
| "loss": 0.334, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.00018683848977691785, | |
| "loss": 0.3342, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 0.0001864838361680247, | |
| "loss": 0.3693, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.00018612481349596405, | |
| "loss": 0.3422, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.00018576143989814523, | |
| "loss": 0.3346, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.00018539373373178123, | |
| "loss": 0.3329, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "eval_loss": 0.3422779440879822, | |
| "eval_runtime": 35.1197, | |
| "eval_samples_per_second": 173.521, | |
| "eval_steps_per_second": 12.415, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.00018502171357296144, | |
| "loss": 0.3485, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.00018464539821571303, | |
| "loss": 0.3383, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 0.00018426480667105177, | |
| "loss": 0.3434, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.00018387995816602137, | |
| "loss": 0.3205, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 0.00018349087214272224, | |
| "loss": 0.3522, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 0.0001830975682573293, | |
| "loss": 0.3416, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 0.00018270006637909907, | |
| "loss": 0.3468, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 0.00018229838658936564, | |
| "loss": 0.3407, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 0.0001818925491805265, | |
| "loss": 0.3198, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 0.00018148257465501716, | |
| "loss": 0.3369, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 0.0001810684837242755, | |
| "loss": 0.3371, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.00018065029730769534, | |
| "loss": 0.3395, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 0.00018022803653156984, | |
| "loss": 0.3418, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 0.000179801722728024, | |
| "loss": 0.346, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 0.00017937137743393694, | |
| "loss": 0.3385, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 0.00017893702238985432, | |
| "loss": 0.3343, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 0.00017849867953888948, | |
| "loss": 0.3146, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 0.00017805637102561516, | |
| "loss": 0.3662, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 0.00017761011919494488, | |
| "loss": 0.3481, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 0.00017715994659100388, | |
| "loss": 0.3447, | |
| "step": 107 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 0.00017670587595599034, | |
| "loss": 0.3379, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 0.0001762479302290265, | |
| "loss": 0.3388, | |
| "step": 109 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 0.0001757861325449997, | |
| "loss": 0.3497, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 0.0001753205062333937, | |
| "loss": 0.3539, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 0.00017485107481711012, | |
| "loss": 0.3571, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 0.00017437786201128004, | |
| "loss": 0.3496, | |
| "step": 113 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 452, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 113, | |
| "total_flos": 5.240873872315646e+18, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |