| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 24, | |
| "global_step": 94, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.010638297872340425, | |
| "grad_norm": 0.10508890450000763, | |
| "learning_rate": 3.0000000000000004e-08, | |
| "loss": 1.2333, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.010638297872340425, | |
| "eval_loss": 1.5895557403564453, | |
| "eval_runtime": 2.9603, | |
| "eval_samples_per_second": 35.469, | |
| "eval_steps_per_second": 17.903, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.02127659574468085, | |
| "grad_norm": 0.12071573734283447, | |
| "learning_rate": 6.000000000000001e-08, | |
| "loss": 1.3986, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.031914893617021274, | |
| "grad_norm": 0.10401547700166702, | |
| "learning_rate": 9e-08, | |
| "loss": 1.2058, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0425531914893617, | |
| "grad_norm": 0.12298297882080078, | |
| "learning_rate": 1.2000000000000002e-07, | |
| "loss": 1.3298, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.05319148936170213, | |
| "grad_norm": 0.1327030211687088, | |
| "learning_rate": 1.5000000000000002e-07, | |
| "loss": 1.4031, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.06382978723404255, | |
| "grad_norm": 0.12036772817373276, | |
| "learning_rate": 1.8e-07, | |
| "loss": 1.5335, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.07446808510638298, | |
| "grad_norm": 0.14979703724384308, | |
| "learning_rate": 2.1000000000000003e-07, | |
| "loss": 1.3224, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0851063829787234, | |
| "grad_norm": 0.13082227110862732, | |
| "learning_rate": 2.4000000000000003e-07, | |
| "loss": 1.4035, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.09574468085106383, | |
| "grad_norm": 0.09265327453613281, | |
| "learning_rate": 2.7e-07, | |
| "loss": 1.1546, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.10638297872340426, | |
| "grad_norm": 0.10080434381961823, | |
| "learning_rate": 3.0000000000000004e-07, | |
| "loss": 1.2652, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.11702127659574468, | |
| "grad_norm": 0.10915568470954895, | |
| "learning_rate": 3.3e-07, | |
| "loss": 1.1438, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.1276595744680851, | |
| "grad_norm": 0.1307348906993866, | |
| "learning_rate": 3.6e-07, | |
| "loss": 1.2821, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.13829787234042554, | |
| "grad_norm": 0.09814529865980148, | |
| "learning_rate": 3.9e-07, | |
| "loss": 1.3905, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.14893617021276595, | |
| "grad_norm": 0.11842218786478043, | |
| "learning_rate": 4.2000000000000006e-07, | |
| "loss": 1.3085, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.1595744680851064, | |
| "grad_norm": 0.10207124054431915, | |
| "learning_rate": 4.5e-07, | |
| "loss": 1.19, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.1702127659574468, | |
| "grad_norm": 0.1150127425789833, | |
| "learning_rate": 4.800000000000001e-07, | |
| "loss": 1.386, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.18085106382978725, | |
| "grad_norm": 0.11641352623701096, | |
| "learning_rate": 5.100000000000001e-07, | |
| "loss": 1.4357, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.19148936170212766, | |
| "grad_norm": 0.15035435557365417, | |
| "learning_rate": 5.4e-07, | |
| "loss": 1.2549, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.20212765957446807, | |
| "grad_norm": 0.0984087809920311, | |
| "learning_rate": 5.7e-07, | |
| "loss": 0.9723, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.2127659574468085, | |
| "grad_norm": 0.11582670360803604, | |
| "learning_rate": 6.000000000000001e-07, | |
| "loss": 1.1402, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.22340425531914893, | |
| "grad_norm": 0.10151291638612747, | |
| "learning_rate": 6.3e-07, | |
| "loss": 1.1533, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.23404255319148937, | |
| "grad_norm": 0.09040653705596924, | |
| "learning_rate": 6.6e-07, | |
| "loss": 1.0438, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.24468085106382978, | |
| "grad_norm": 0.10666865110397339, | |
| "learning_rate": 6.900000000000001e-07, | |
| "loss": 1.3195, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.2553191489361702, | |
| "grad_norm": 0.12293969094753265, | |
| "learning_rate": 7.2e-07, | |
| "loss": 1.7286, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.2553191489361702, | |
| "eval_loss": 1.5890721082687378, | |
| "eval_runtime": 2.9908, | |
| "eval_samples_per_second": 35.108, | |
| "eval_steps_per_second": 17.721, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.26595744680851063, | |
| "grad_norm": 0.11052387952804565, | |
| "learning_rate": 7.5e-07, | |
| "loss": 1.1413, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.2765957446808511, | |
| "grad_norm": 0.09323304146528244, | |
| "learning_rate": 7.8e-07, | |
| "loss": 1.2906, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.2872340425531915, | |
| "grad_norm": 0.12542971968650818, | |
| "learning_rate": 8.100000000000001e-07, | |
| "loss": 1.2809, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.2978723404255319, | |
| "grad_norm": 0.1080215722322464, | |
| "learning_rate": 8.400000000000001e-07, | |
| "loss": 1.2735, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.30851063829787234, | |
| "grad_norm": 0.11304887384176254, | |
| "learning_rate": 8.699999999999999e-07, | |
| "loss": 1.3767, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.3191489361702128, | |
| "grad_norm": 0.15650290250778198, | |
| "learning_rate": 9e-07, | |
| "loss": 1.4713, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.32978723404255317, | |
| "grad_norm": 0.12139321118593216, | |
| "learning_rate": 9.3e-07, | |
| "loss": 1.3479, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.3404255319148936, | |
| "grad_norm": 0.10141867399215698, | |
| "learning_rate": 9.600000000000001e-07, | |
| "loss": 1.1512, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.35106382978723405, | |
| "grad_norm": 0.15045498311519623, | |
| "learning_rate": 9.9e-07, | |
| "loss": 1.5059, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.3617021276595745, | |
| "grad_norm": 0.12956956028938293, | |
| "learning_rate": 1.0200000000000002e-06, | |
| "loss": 1.2849, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.3723404255319149, | |
| "grad_norm": 0.12963886559009552, | |
| "learning_rate": 1.05e-06, | |
| "loss": 1.5025, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.3829787234042553, | |
| "grad_norm": 0.11268144845962524, | |
| "learning_rate": 1.08e-06, | |
| "loss": 1.2987, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.39361702127659576, | |
| "grad_norm": 0.12941108644008636, | |
| "learning_rate": 1.11e-06, | |
| "loss": 1.3432, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.40425531914893614, | |
| "grad_norm": 0.11319927126169205, | |
| "learning_rate": 1.14e-06, | |
| "loss": 1.1665, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.4148936170212766, | |
| "grad_norm": 0.11748912930488586, | |
| "learning_rate": 1.17e-06, | |
| "loss": 1.423, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.425531914893617, | |
| "grad_norm": 0.11666171997785568, | |
| "learning_rate": 1.2000000000000002e-06, | |
| "loss": 1.4391, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.43617021276595747, | |
| "grad_norm": 0.12274409830570221, | |
| "learning_rate": 1.2299999999999999e-06, | |
| "loss": 1.4865, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.44680851063829785, | |
| "grad_norm": 0.09922561049461365, | |
| "learning_rate": 1.26e-06, | |
| "loss": 1.2193, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.4574468085106383, | |
| "grad_norm": 0.12003930658102036, | |
| "learning_rate": 1.29e-06, | |
| "loss": 1.3024, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.46808510638297873, | |
| "grad_norm": 0.12094161659479141, | |
| "learning_rate": 1.32e-06, | |
| "loss": 1.4654, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.4787234042553192, | |
| "grad_norm": 0.12934774160385132, | |
| "learning_rate": 1.35e-06, | |
| "loss": 1.3713, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.48936170212765956, | |
| "grad_norm": 0.09754550457000732, | |
| "learning_rate": 1.3800000000000001e-06, | |
| "loss": 1.1991, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.11549004167318344, | |
| "learning_rate": 1.41e-06, | |
| "loss": 1.4859, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.5106382978723404, | |
| "grad_norm": 0.12035688012838364, | |
| "learning_rate": 1.44e-06, | |
| "loss": 1.2823, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.5106382978723404, | |
| "eval_loss": 1.5874611139297485, | |
| "eval_runtime": 2.9787, | |
| "eval_samples_per_second": 35.251, | |
| "eval_steps_per_second": 17.793, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.5212765957446809, | |
| "grad_norm": 0.12258938699960709, | |
| "learning_rate": 1.4700000000000001e-06, | |
| "loss": 1.2966, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.5319148936170213, | |
| "grad_norm": 0.12217017263174057, | |
| "learning_rate": 1.5e-06, | |
| "loss": 1.2988, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.5425531914893617, | |
| "grad_norm": 0.12793436646461487, | |
| "learning_rate": 1.53e-06, | |
| "loss": 1.4233, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.5531914893617021, | |
| "grad_norm": 0.11145548522472382, | |
| "learning_rate": 1.56e-06, | |
| "loss": 1.2792, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.5638297872340425, | |
| "grad_norm": 0.13195408880710602, | |
| "learning_rate": 1.59e-06, | |
| "loss": 1.4481, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.574468085106383, | |
| "grad_norm": 0.10663347691297531, | |
| "learning_rate": 1.6200000000000002e-06, | |
| "loss": 1.2331, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.5851063829787234, | |
| "grad_norm": 0.10975392907857895, | |
| "learning_rate": 1.65e-06, | |
| "loss": 1.1839, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.5957446808510638, | |
| "grad_norm": 0.13139477372169495, | |
| "learning_rate": 1.6800000000000002e-06, | |
| "loss": 1.436, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.6063829787234043, | |
| "grad_norm": 0.0924743041396141, | |
| "learning_rate": 1.71e-06, | |
| "loss": 1.1291, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.6170212765957447, | |
| "grad_norm": 0.12475109100341797, | |
| "learning_rate": 1.7399999999999999e-06, | |
| "loss": 1.3444, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.6276595744680851, | |
| "grad_norm": 0.08960220962762833, | |
| "learning_rate": 1.77e-06, | |
| "loss": 1.2511, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.6382978723404256, | |
| "grad_norm": 0.09909304976463318, | |
| "learning_rate": 1.8e-06, | |
| "loss": 1.1281, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.648936170212766, | |
| "grad_norm": 0.11598234623670578, | |
| "learning_rate": 1.83e-06, | |
| "loss": 1.486, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.6595744680851063, | |
| "grad_norm": 0.1404409557580948, | |
| "learning_rate": 1.86e-06, | |
| "loss": 1.4557, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.6702127659574468, | |
| "grad_norm": 0.11349129676818848, | |
| "learning_rate": 1.8900000000000001e-06, | |
| "loss": 1.3969, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.6808510638297872, | |
| "grad_norm": 0.10858353972434998, | |
| "learning_rate": 1.9200000000000003e-06, | |
| "loss": 1.3515, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.6914893617021277, | |
| "grad_norm": 0.11054569482803345, | |
| "learning_rate": 1.95e-06, | |
| "loss": 1.248, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.7021276595744681, | |
| "grad_norm": 0.11826737225055695, | |
| "learning_rate": 1.98e-06, | |
| "loss": 1.3439, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.7127659574468085, | |
| "grad_norm": 0.12291310727596283, | |
| "learning_rate": 2.0100000000000002e-06, | |
| "loss": 1.4226, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.723404255319149, | |
| "grad_norm": 0.1383126825094223, | |
| "learning_rate": 2.0400000000000004e-06, | |
| "loss": 1.194, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.7340425531914894, | |
| "grad_norm": 0.10981890559196472, | |
| "learning_rate": 2.07e-06, | |
| "loss": 1.2502, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.7446808510638298, | |
| "grad_norm": 0.10639657825231552, | |
| "learning_rate": 2.1e-06, | |
| "loss": 1.2262, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.7553191489361702, | |
| "grad_norm": 0.46177396178245544, | |
| "learning_rate": 2.13e-06, | |
| "loss": 5.063, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.7659574468085106, | |
| "grad_norm": 0.11061427742242813, | |
| "learning_rate": 2.16e-06, | |
| "loss": 1.3856, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.7659574468085106, | |
| "eval_loss": 1.5843762159347534, | |
| "eval_runtime": 2.9943, | |
| "eval_samples_per_second": 35.066, | |
| "eval_steps_per_second": 17.7, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.776595744680851, | |
| "grad_norm": 0.13233324885368347, | |
| "learning_rate": 2.19e-06, | |
| "loss": 1.526, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.7872340425531915, | |
| "grad_norm": 0.14500053226947784, | |
| "learning_rate": 2.22e-06, | |
| "loss": 1.2554, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.7978723404255319, | |
| "grad_norm": 0.10629269480705261, | |
| "learning_rate": 2.25e-06, | |
| "loss": 1.2917, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.8085106382978723, | |
| "grad_norm": 0.09674028307199478, | |
| "learning_rate": 2.28e-06, | |
| "loss": 1.2169, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.8191489361702128, | |
| "grad_norm": 0.11965947598218918, | |
| "learning_rate": 2.31e-06, | |
| "loss": 1.2445, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 0.8297872340425532, | |
| "grad_norm": 0.10441229492425919, | |
| "learning_rate": 2.34e-06, | |
| "loss": 1.2392, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.8404255319148937, | |
| "grad_norm": 0.13795869052410126, | |
| "learning_rate": 2.37e-06, | |
| "loss": 1.5335, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 0.851063829787234, | |
| "grad_norm": 0.10789927840232849, | |
| "learning_rate": 2.4000000000000003e-06, | |
| "loss": 1.2309, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.8617021276595744, | |
| "grad_norm": 0.10697130113840103, | |
| "learning_rate": 2.43e-06, | |
| "loss": 1.276, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.8723404255319149, | |
| "grad_norm": 0.11484125256538391, | |
| "learning_rate": 2.4599999999999997e-06, | |
| "loss": 1.2163, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 0.8829787234042553, | |
| "grad_norm": 0.09692610800266266, | |
| "learning_rate": 2.49e-06, | |
| "loss": 1.1554, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 0.8936170212765957, | |
| "grad_norm": 0.10697747021913528, | |
| "learning_rate": 2.52e-06, | |
| "loss": 1.211, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.9042553191489362, | |
| "grad_norm": 0.10578318685293198, | |
| "learning_rate": 2.55e-06, | |
| "loss": 1.2634, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.9148936170212766, | |
| "grad_norm": 0.10587752610445023, | |
| "learning_rate": 2.58e-06, | |
| "loss": 1.369, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 0.925531914893617, | |
| "grad_norm": 0.11430489271879196, | |
| "learning_rate": 2.61e-06, | |
| "loss": 1.3048, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.9361702127659575, | |
| "grad_norm": 0.11116154491901398, | |
| "learning_rate": 2.64e-06, | |
| "loss": 1.1883, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.9468085106382979, | |
| "grad_norm": 0.12686476111412048, | |
| "learning_rate": 2.6700000000000003e-06, | |
| "loss": 1.2605, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 0.9574468085106383, | |
| "grad_norm": 0.10976041853427887, | |
| "learning_rate": 2.7e-06, | |
| "loss": 1.2345, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.9680851063829787, | |
| "grad_norm": 0.12391550838947296, | |
| "learning_rate": 2.73e-06, | |
| "loss": 1.2233, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 0.9787234042553191, | |
| "grad_norm": 0.1277073174715042, | |
| "learning_rate": 2.7600000000000003e-06, | |
| "loss": 1.2437, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 0.9893617021276596, | |
| "grad_norm": 0.10421616584062576, | |
| "learning_rate": 2.7900000000000004e-06, | |
| "loss": 1.1432, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.12696842849254608, | |
| "learning_rate": 2.82e-06, | |
| "loss": 1.3558, | |
| "step": 94 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 376, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 94, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3164980962656256.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |