| { | |
| "best_metric": 0.2518707811832428, | |
| "best_model_checkpoint": "/shared/user/experiments/comps/finnlp/task2/llama3_8b_fft_sent_w_summ_01/checkpoint-160", | |
| "epoch": 3.747832939322301, | |
| "eval_steps": 20, | |
| "global_step": 300, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01260835303388495, | |
| "grad_norm": 27.375, | |
| "learning_rate": 1.25e-06, | |
| "loss": 0.7607, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.01260835303388495, | |
| "eval_loss": 0.9510096311569214, | |
| "eval_runtime": 134.2946, | |
| "eval_samples_per_second": 1.467, | |
| "eval_steps_per_second": 1.467, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0252167060677699, | |
| "grad_norm": 30.0, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.8758, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.037825059101654845, | |
| "grad_norm": 35.5, | |
| "learning_rate": 3.7500000000000005e-06, | |
| "loss": 1.0098, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0504334121355398, | |
| "grad_norm": 23.25, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6727, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.06304176516942474, | |
| "grad_norm": 10.6875, | |
| "learning_rate": 6.25e-06, | |
| "loss": 0.4857, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.07565011820330969, | |
| "grad_norm": 10.0, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.4054, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.08825847123719464, | |
| "grad_norm": 7.84375, | |
| "learning_rate": 8.750000000000001e-06, | |
| "loss": 0.3342, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.1008668242710796, | |
| "grad_norm": 7.09375, | |
| "learning_rate": 1e-05, | |
| "loss": 0.3769, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.11347517730496454, | |
| "grad_norm": 4.1875, | |
| "learning_rate": 9.999739903471488e-06, | |
| "loss": 0.3558, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.12608353033884948, | |
| "grad_norm": 7.3125, | |
| "learning_rate": 9.998959640946033e-06, | |
| "loss": 0.4658, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.13869188337273444, | |
| "grad_norm": 5.0625, | |
| "learning_rate": 9.997659293601066e-06, | |
| "loss": 0.3392, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.15130023640661938, | |
| "grad_norm": 1.5625, | |
| "learning_rate": 9.995838996722916e-06, | |
| "loss": 0.2612, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.16390858944050432, | |
| "grad_norm": 2.078125, | |
| "learning_rate": 9.993498939692744e-06, | |
| "loss": 0.2597, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.1765169424743893, | |
| "grad_norm": 2.09375, | |
| "learning_rate": 9.990639365966835e-06, | |
| "loss": 0.3113, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.18912529550827423, | |
| "grad_norm": 1.8671875, | |
| "learning_rate": 9.987260573051268e-06, | |
| "loss": 0.2713, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.2017336485421592, | |
| "grad_norm": 4.875, | |
| "learning_rate": 9.983362912470967e-06, | |
| "loss": 0.2834, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.21434200157604413, | |
| "grad_norm": 1.625, | |
| "learning_rate": 9.978946789733126e-06, | |
| "loss": 0.2198, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.22695035460992907, | |
| "grad_norm": 5.0625, | |
| "learning_rate": 9.97401266428502e-06, | |
| "loss": 0.2978, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.23955870764381404, | |
| "grad_norm": 1.375, | |
| "learning_rate": 9.968561049466214e-06, | |
| "loss": 0.236, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.25216706067769895, | |
| "grad_norm": 6.75, | |
| "learning_rate": 9.96259251245514e-06, | |
| "loss": 0.277, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.25216706067769895, | |
| "eval_loss": 0.27656540274620056, | |
| "eval_runtime": 132.7922, | |
| "eval_samples_per_second": 1.484, | |
| "eval_steps_per_second": 1.484, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.2647754137115839, | |
| "grad_norm": 2.59375, | |
| "learning_rate": 9.9561076742101e-06, | |
| "loss": 0.2649, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.2773837667454689, | |
| "grad_norm": 2.625, | |
| "learning_rate": 9.949107209404664e-06, | |
| "loss": 0.2352, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.2899921197793538, | |
| "grad_norm": 3.84375, | |
| "learning_rate": 9.941591846357467e-06, | |
| "loss": 0.3047, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.30260047281323876, | |
| "grad_norm": 1.5390625, | |
| "learning_rate": 9.933562366956445e-06, | |
| "loss": 0.228, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.31520882584712373, | |
| "grad_norm": 1.46875, | |
| "learning_rate": 9.925019606577486e-06, | |
| "loss": 0.2075, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.32781717888100864, | |
| "grad_norm": 1.328125, | |
| "learning_rate": 9.915964453997516e-06, | |
| "loss": 0.2222, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.3404255319148936, | |
| "grad_norm": 1.5, | |
| "learning_rate": 9.906397851302036e-06, | |
| "loss": 0.2003, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.3530338849487786, | |
| "grad_norm": 3.078125, | |
| "learning_rate": 9.896320793787106e-06, | |
| "loss": 0.2966, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.3656422379826635, | |
| "grad_norm": 1.1953125, | |
| "learning_rate": 9.885734329855798e-06, | |
| "loss": 0.2072, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.37825059101654845, | |
| "grad_norm": 1.484375, | |
| "learning_rate": 9.874639560909118e-06, | |
| "loss": 0.2204, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.3908589440504334, | |
| "grad_norm": 1.4375, | |
| "learning_rate": 9.863037641231424e-06, | |
| "loss": 0.278, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.4034672970843184, | |
| "grad_norm": 1.734375, | |
| "learning_rate": 9.850929777870324e-06, | |
| "loss": 0.1975, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.4160756501182033, | |
| "grad_norm": 1.5625, | |
| "learning_rate": 9.838317230511111e-06, | |
| "loss": 0.1582, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.42868400315208827, | |
| "grad_norm": 1.265625, | |
| "learning_rate": 9.8252013113457e-06, | |
| "loss": 0.2326, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.44129235618597323, | |
| "grad_norm": 1.21875, | |
| "learning_rate": 9.811583384936108e-06, | |
| "loss": 0.2214, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.45390070921985815, | |
| "grad_norm": 1.421875, | |
| "learning_rate": 9.797464868072489e-06, | |
| "loss": 0.2355, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.4665090622537431, | |
| "grad_norm": 1.171875, | |
| "learning_rate": 9.782847229625729e-06, | |
| "loss": 0.1939, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.4791174152876281, | |
| "grad_norm": 1.171875, | |
| "learning_rate": 9.767731990394638e-06, | |
| "loss": 0.195, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.491725768321513, | |
| "grad_norm": 1.8203125, | |
| "learning_rate": 9.752120722947717e-06, | |
| "loss": 0.2297, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.5043341213553979, | |
| "grad_norm": 1.46875, | |
| "learning_rate": 9.736015051459551e-06, | |
| "loss": 0.1991, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5043341213553979, | |
| "eval_loss": 0.2646142840385437, | |
| "eval_runtime": 132.7773, | |
| "eval_samples_per_second": 1.484, | |
| "eval_steps_per_second": 1.484, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5169424743892829, | |
| "grad_norm": 1.75, | |
| "learning_rate": 9.719416651541839e-06, | |
| "loss": 0.2207, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.5295508274231678, | |
| "grad_norm": 1.53125, | |
| "learning_rate": 9.702327250069058e-06, | |
| "loss": 0.2224, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.5421591804570528, | |
| "grad_norm": 3.796875, | |
| "learning_rate": 9.68474862499881e-06, | |
| "loss": 0.2688, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.5547675334909378, | |
| "grad_norm": 1.546875, | |
| "learning_rate": 9.666682605186834e-06, | |
| "loss": 0.2435, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.5673758865248227, | |
| "grad_norm": 1.3359375, | |
| "learning_rate": 9.648131070196749e-06, | |
| "loss": 0.2405, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.5799842395587076, | |
| "grad_norm": 5.8125, | |
| "learning_rate": 9.62909595010449e-06, | |
| "loss": 0.2553, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.5925925925925926, | |
| "grad_norm": 1.4140625, | |
| "learning_rate": 9.609579225297524e-06, | |
| "loss": 0.2323, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.6052009456264775, | |
| "grad_norm": 1.46875, | |
| "learning_rate": 9.589582926268798e-06, | |
| "loss": 0.2531, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.6178092986603625, | |
| "grad_norm": 1.1640625, | |
| "learning_rate": 9.569109133405495e-06, | |
| "loss": 0.229, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.6304176516942475, | |
| "grad_norm": 1.28125, | |
| "learning_rate": 9.548159976772593e-06, | |
| "loss": 0.2168, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.6430260047281324, | |
| "grad_norm": 1.1640625, | |
| "learning_rate": 9.526737635891262e-06, | |
| "loss": 0.2164, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.6556343577620173, | |
| "grad_norm": 1.234375, | |
| "learning_rate": 9.504844339512096e-06, | |
| "loss": 0.1924, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.6682427107959023, | |
| "grad_norm": 1.6484375, | |
| "learning_rate": 9.482482365383254e-06, | |
| "loss": 0.2351, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.6808510638297872, | |
| "grad_norm": 1.234375, | |
| "learning_rate": 9.45965404001347e-06, | |
| "loss": 0.2145, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.6934594168636722, | |
| "grad_norm": 3.09375, | |
| "learning_rate": 9.436361738430016e-06, | |
| "loss": 0.2935, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.7060677698975572, | |
| "grad_norm": 1.6484375, | |
| "learning_rate": 9.412607883931608e-06, | |
| "loss": 0.2509, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.7186761229314421, | |
| "grad_norm": 1.28125, | |
| "learning_rate": 9.388394947836278e-06, | |
| "loss": 0.2136, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.731284475965327, | |
| "grad_norm": 1.3359375, | |
| "learning_rate": 9.363725449224281e-06, | |
| "loss": 0.2224, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.7438928289992119, | |
| "grad_norm": 1.140625, | |
| "learning_rate": 9.338601954675995e-06, | |
| "loss": 0.2303, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.7565011820330969, | |
| "grad_norm": 1.4453125, | |
| "learning_rate": 9.313027078004903e-06, | |
| "loss": 0.2292, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.7565011820330969, | |
| "eval_loss": 0.26008567214012146, | |
| "eval_runtime": 134.4418, | |
| "eval_samples_per_second": 1.465, | |
| "eval_steps_per_second": 1.465, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.7691095350669819, | |
| "grad_norm": 1.3046875, | |
| "learning_rate": 9.287003479985667e-06, | |
| "loss": 0.2092, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.7817178881008668, | |
| "grad_norm": 1.3046875, | |
| "learning_rate": 9.260533868077283e-06, | |
| "loss": 0.2087, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.7943262411347518, | |
| "grad_norm": 1.1796875, | |
| "learning_rate": 9.233620996141421e-06, | |
| "loss": 0.2079, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.8069345941686368, | |
| "grad_norm": 3.921875, | |
| "learning_rate": 9.206267664155906e-06, | |
| "loss": 0.2365, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.8195429472025216, | |
| "grad_norm": 1.484375, | |
| "learning_rate": 9.178476717923415e-06, | |
| "loss": 0.233, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.8321513002364066, | |
| "grad_norm": 1.6640625, | |
| "learning_rate": 9.150251048775403e-06, | |
| "loss": 0.2413, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.8447596532702916, | |
| "grad_norm": 1.21875, | |
| "learning_rate": 9.121593593271297e-06, | |
| "loss": 0.2212, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.8573680063041765, | |
| "grad_norm": 1.296875, | |
| "learning_rate": 9.092507332892968e-06, | |
| "loss": 0.236, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.8699763593380615, | |
| "grad_norm": 1.3046875, | |
| "learning_rate": 9.062995293734562e-06, | |
| "loss": 0.2259, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.8825847123719465, | |
| "grad_norm": 1.15625, | |
| "learning_rate": 9.033060546187651e-06, | |
| "loss": 0.201, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.8951930654058313, | |
| "grad_norm": 1.6328125, | |
| "learning_rate": 9.002706204621802e-06, | |
| "loss": 0.2318, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.9078014184397163, | |
| "grad_norm": 1.0703125, | |
| "learning_rate": 8.971935427060563e-06, | |
| "loss": 0.1886, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.9204097714736013, | |
| "grad_norm": 1.0546875, | |
| "learning_rate": 8.940751414852904e-06, | |
| "loss": 0.1883, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.9330181245074862, | |
| "grad_norm": 1.53125, | |
| "learning_rate": 8.90915741234015e-06, | |
| "loss": 0.2578, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.9456264775413712, | |
| "grad_norm": 1.1796875, | |
| "learning_rate": 8.877156706518453e-06, | |
| "loss": 0.1865, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.9582348305752562, | |
| "grad_norm": 1.171875, | |
| "learning_rate": 8.84475262669681e-06, | |
| "loss": 0.2116, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.970843183609141, | |
| "grad_norm": 1.1171875, | |
| "learning_rate": 8.811948544150693e-06, | |
| "loss": 0.1951, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 0.983451536643026, | |
| "grad_norm": 3.25, | |
| "learning_rate": 8.778747871771293e-06, | |
| "loss": 0.2327, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.996059889676911, | |
| "grad_norm": 1.2109375, | |
| "learning_rate": 8.745154063710464e-06, | |
| "loss": 0.2052, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 1.0086682427107958, | |
| "grad_norm": 1.0546875, | |
| "learning_rate": 8.71117061502135e-06, | |
| "loss": 0.188, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.0086682427107958, | |
| "eval_loss": 0.253794401884079, | |
| "eval_runtime": 133.0721, | |
| "eval_samples_per_second": 1.48, | |
| "eval_steps_per_second": 1.48, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.0102442868400314, | |
| "grad_norm": 1.421875, | |
| "learning_rate": 8.676801061294764e-06, | |
| "loss": 0.2204, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.0228526398739164, | |
| "grad_norm": 2.8125, | |
| "learning_rate": 8.642048978291347e-06, | |
| "loss": 0.2404, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 1.0354609929078014, | |
| "grad_norm": 1.5078125, | |
| "learning_rate": 8.60691798156956e-06, | |
| "loss": 0.1494, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 1.0480693459416863, | |
| "grad_norm": 1.1640625, | |
| "learning_rate": 8.571411726109518e-06, | |
| "loss": 0.1717, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.0606776989755713, | |
| "grad_norm": 1.1953125, | |
| "learning_rate": 8.535533905932739e-06, | |
| "loss": 0.1789, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 1.0732860520094563, | |
| "grad_norm": 1.078125, | |
| "learning_rate": 8.49928825371781e-06, | |
| "loss": 0.152, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 1.0858944050433412, | |
| "grad_norm": 1.1796875, | |
| "learning_rate": 8.46267854041206e-06, | |
| "loss": 0.1999, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 1.0985027580772262, | |
| "grad_norm": 1.203125, | |
| "learning_rate": 8.425708574839221e-06, | |
| "loss": 0.1751, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 1.1111111111111112, | |
| "grad_norm": 1.2421875, | |
| "learning_rate": 8.388382203303181e-06, | |
| "loss": 0.1942, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 1.1237194641449961, | |
| "grad_norm": 1.25, | |
| "learning_rate": 8.3507033091878e-06, | |
| "loss": 0.1744, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.136327817178881, | |
| "grad_norm": 1.71875, | |
| "learning_rate": 8.312675812552898e-06, | |
| "loss": 0.1717, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 1.148936170212766, | |
| "grad_norm": 1.234375, | |
| "learning_rate": 8.274303669726427e-06, | |
| "loss": 0.1906, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 1.161544523246651, | |
| "grad_norm": 1.1875, | |
| "learning_rate": 8.235590872892837e-06, | |
| "loss": 0.1609, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 1.1741528762805358, | |
| "grad_norm": 1.2578125, | |
| "learning_rate": 8.196541449677758e-06, | |
| "loss": 0.1837, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 1.1867612293144207, | |
| "grad_norm": 1.3828125, | |
| "learning_rate": 8.157159462728956e-06, | |
| "loss": 0.187, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 1.1993695823483057, | |
| "grad_norm": 1.140625, | |
| "learning_rate": 8.117449009293668e-06, | |
| "loss": 0.1832, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.2119779353821907, | |
| "grad_norm": 1.3203125, | |
| "learning_rate": 8.077414220792328e-06, | |
| "loss": 0.1785, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 1.2245862884160756, | |
| "grad_norm": 1.375, | |
| "learning_rate": 8.03705926238874e-06, | |
| "loss": 0.196, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 1.2371946414499606, | |
| "grad_norm": 1.15625, | |
| "learning_rate": 7.996388332556735e-06, | |
| "loss": 0.186, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 1.2498029944838456, | |
| "grad_norm": 1.0390625, | |
| "learning_rate": 7.955405662643384e-06, | |
| "loss": 0.1689, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.2498029944838456, | |
| "eval_loss": 0.253927618265152, | |
| "eval_runtime": 132.9337, | |
| "eval_samples_per_second": 1.482, | |
| "eval_steps_per_second": 1.482, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.2624113475177305, | |
| "grad_norm": 1.4765625, | |
| "learning_rate": 7.914115516428751e-06, | |
| "loss": 0.1599, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 1.2750197005516155, | |
| "grad_norm": 1.71875, | |
| "learning_rate": 7.872522189682318e-06, | |
| "loss": 0.1776, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 1.2876280535855005, | |
| "grad_norm": 1.2421875, | |
| "learning_rate": 7.830630009716038e-06, | |
| "loss": 0.1825, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 1.3002364066193852, | |
| "grad_norm": 3.09375, | |
| "learning_rate": 7.788443334934148e-06, | |
| "loss": 0.1773, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 1.3128447596532702, | |
| "grad_norm": 1.4375, | |
| "learning_rate": 7.745966554379708e-06, | |
| "loss": 0.1818, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 1.3254531126871552, | |
| "grad_norm": 1.234375, | |
| "learning_rate": 7.703204087277989e-06, | |
| "loss": 0.2064, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 1.3380614657210401, | |
| "grad_norm": 2.453125, | |
| "learning_rate": 7.660160382576683e-06, | |
| "loss": 0.2058, | |
| "step": 107 | |
| }, | |
| { | |
| "epoch": 1.350669818754925, | |
| "grad_norm": 1.140625, | |
| "learning_rate": 7.616839918483061e-06, | |
| "loss": 0.1958, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 1.36327817178881, | |
| "grad_norm": 1.234375, | |
| "learning_rate": 7.573247201998051e-06, | |
| "loss": 0.1887, | |
| "step": 109 | |
| }, | |
| { | |
| "epoch": 1.375886524822695, | |
| "grad_norm": 1.15625, | |
| "learning_rate": 7.529386768447342e-06, | |
| "loss": 0.1763, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.38849487785658, | |
| "grad_norm": 1.2265625, | |
| "learning_rate": 7.485263181009539e-06, | |
| "loss": 0.2287, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 1.401103230890465, | |
| "grad_norm": 1.578125, | |
| "learning_rate": 7.440881030241407e-06, | |
| "loss": 0.2117, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 1.41371158392435, | |
| "grad_norm": 1.5, | |
| "learning_rate": 7.396244933600285e-06, | |
| "loss": 0.1882, | |
| "step": 113 | |
| }, | |
| { | |
| "epoch": 1.426319936958235, | |
| "grad_norm": 2.40625, | |
| "learning_rate": 7.351359534963684e-06, | |
| "loss": 0.1953, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 1.4389282899921199, | |
| "grad_norm": 4.53125, | |
| "learning_rate": 7.306229504146154e-06, | |
| "loss": 0.1932, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 1.4515366430260048, | |
| "grad_norm": 1.1171875, | |
| "learning_rate": 7.260859536413429e-06, | |
| "loss": 0.2039, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 1.4641449960598898, | |
| "grad_norm": 1.3671875, | |
| "learning_rate": 7.215254351993957e-06, | |
| "loss": 0.1908, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 1.4767533490937748, | |
| "grad_norm": 1.265625, | |
| "learning_rate": 7.169418695587791e-06, | |
| "loss": 0.1862, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 1.4893617021276595, | |
| "grad_norm": 1.2265625, | |
| "learning_rate": 7.123357335872982e-06, | |
| "loss": 0.161, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 1.5019700551615445, | |
| "grad_norm": 1.4765625, | |
| "learning_rate": 7.0770750650094335e-06, | |
| "loss": 0.202, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.5019700551615445, | |
| "eval_loss": 0.2536303997039795, | |
| "eval_runtime": 132.5446, | |
| "eval_samples_per_second": 1.486, | |
| "eval_steps_per_second": 1.486, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.5145784081954294, | |
| "grad_norm": 1.1171875, | |
| "learning_rate": 7.0305766981403365e-06, | |
| "loss": 0.2026, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 1.5271867612293144, | |
| "grad_norm": 2.515625, | |
| "learning_rate": 6.983867072891213e-06, | |
| "loss": 0.1875, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.5397951142631994, | |
| "grad_norm": 3.953125, | |
| "learning_rate": 6.936951048866616e-06, | |
| "loss": 0.1899, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 1.5524034672970843, | |
| "grad_norm": 1.109375, | |
| "learning_rate": 6.889833507144534e-06, | |
| "loss": 0.1788, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 1.5650118203309693, | |
| "grad_norm": 1.1484375, | |
| "learning_rate": 6.842519349768582e-06, | |
| "loss": 0.2098, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.577620173364854, | |
| "grad_norm": 1.5546875, | |
| "learning_rate": 6.7950134992379935e-06, | |
| "loss": 0.2224, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 1.590228526398739, | |
| "grad_norm": 2.1875, | |
| "learning_rate": 6.747320897995493e-06, | |
| "loss": 0.1832, | |
| "step": 127 | |
| }, | |
| { | |
| "epoch": 1.602836879432624, | |
| "grad_norm": 1.4453125, | |
| "learning_rate": 6.699446507913083e-06, | |
| "loss": 0.1894, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 1.615445232466509, | |
| "grad_norm": 1.296875, | |
| "learning_rate": 6.651395309775837e-06, | |
| "loss": 0.1822, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 1.628053585500394, | |
| "grad_norm": 2.875, | |
| "learning_rate": 6.6031723027636775e-06, | |
| "loss": 0.202, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.6406619385342789, | |
| "grad_norm": 1.46875, | |
| "learning_rate": 6.554782503931298e-06, | |
| "loss": 0.1728, | |
| "step": 131 | |
| }, | |
| { | |
| "epoch": 1.6532702915681639, | |
| "grad_norm": 1.5078125, | |
| "learning_rate": 6.5062309476861714e-06, | |
| "loss": 0.1936, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 1.6658786446020488, | |
| "grad_norm": 1.0859375, | |
| "learning_rate": 6.457522685264793e-06, | |
| "loss": 0.1848, | |
| "step": 133 | |
| }, | |
| { | |
| "epoch": 1.6784869976359338, | |
| "grad_norm": 1.421875, | |
| "learning_rate": 6.408662784207149e-06, | |
| "loss": 0.1981, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 1.6910953506698188, | |
| "grad_norm": 1.203125, | |
| "learning_rate": 6.359656327829498e-06, | |
| "loss": 0.1723, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 1.7037037037037037, | |
| "grad_norm": 1.640625, | |
| "learning_rate": 6.310508414695511e-06, | |
| "loss": 0.1761, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 1.7163120567375887, | |
| "grad_norm": 1.7265625, | |
| "learning_rate": 6.261224158085826e-06, | |
| "loss": 0.1953, | |
| "step": 137 | |
| }, | |
| { | |
| "epoch": 1.7289204097714737, | |
| "grad_norm": 1.171875, | |
| "learning_rate": 6.211808685466063e-06, | |
| "loss": 0.1978, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 1.7415287628053586, | |
| "grad_norm": 1.265625, | |
| "learning_rate": 6.162267137953374e-06, | |
| "loss": 0.1509, | |
| "step": 139 | |
| }, | |
| { | |
| "epoch": 1.7541371158392436, | |
| "grad_norm": 1.1640625, | |
| "learning_rate": 6.112604669781572e-06, | |
| "loss": 0.1807, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.7541371158392436, | |
| "eval_loss": 0.2524150013923645, | |
| "eval_runtime": 133.0086, | |
| "eval_samples_per_second": 1.481, | |
| "eval_steps_per_second": 1.481, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.7667454688731286, | |
| "grad_norm": 1.3046875, | |
| "learning_rate": 6.062826447764883e-06, | |
| "loss": 0.1879, | |
| "step": 141 | |
| }, | |
| { | |
| "epoch": 1.7793538219070135, | |
| "grad_norm": 1.4609375, | |
| "learning_rate": 6.012937650760406e-06, | |
| "loss": 0.1709, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 1.7919621749408985, | |
| "grad_norm": 1.5234375, | |
| "learning_rate": 5.962943469129303e-06, | |
| "loss": 0.1732, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 1.8045705279747835, | |
| "grad_norm": 1.359375, | |
| "learning_rate": 5.91284910419681e-06, | |
| "loss": 0.1724, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 1.8171788810086682, | |
| "grad_norm": 1.3828125, | |
| "learning_rate": 5.862659767711094e-06, | |
| "loss": 0.2049, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 1.8297872340425532, | |
| "grad_norm": 1.28125, | |
| "learning_rate": 5.812380681301031e-06, | |
| "loss": 0.1809, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 1.8423955870764381, | |
| "grad_norm": 1.2109375, | |
| "learning_rate": 5.762017075932952e-06, | |
| "loss": 0.1694, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 1.855003940110323, | |
| "grad_norm": 1.28125, | |
| "learning_rate": 5.711574191366427e-06, | |
| "loss": 0.2312, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 1.867612293144208, | |
| "grad_norm": 1.171875, | |
| "learning_rate": 5.66105727560912e-06, | |
| "loss": 0.172, | |
| "step": 149 | |
| }, | |
| { | |
| "epoch": 1.880220646178093, | |
| "grad_norm": 1.2890625, | |
| "learning_rate": 5.6104715843708e-06, | |
| "loss": 0.2074, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.8928289992119778, | |
| "grad_norm": 1.1796875, | |
| "learning_rate": 5.559822380516539e-06, | |
| "loss": 0.1849, | |
| "step": 151 | |
| }, | |
| { | |
| "epoch": 1.9054373522458627, | |
| "grad_norm": 1.8203125, | |
| "learning_rate": 5.509114933519179e-06, | |
| "loss": 0.1635, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 1.9180457052797477, | |
| "grad_norm": 1.140625, | |
| "learning_rate": 5.458354518911086e-06, | |
| "loss": 0.1956, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 1.9306540583136327, | |
| "grad_norm": 1.3671875, | |
| "learning_rate": 5.4075464177353165e-06, | |
| "loss": 0.1832, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 1.9432624113475176, | |
| "grad_norm": 1.1640625, | |
| "learning_rate": 5.356695915996162e-06, | |
| "loss": 0.1713, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 1.9558707643814026, | |
| "grad_norm": 1.2890625, | |
| "learning_rate": 5.3058083041092145e-06, | |
| "loss": 0.2112, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 1.9684791174152876, | |
| "grad_norm": 1.375, | |
| "learning_rate": 5.25488887635095e-06, | |
| "loss": 0.1613, | |
| "step": 157 | |
| }, | |
| { | |
| "epoch": 1.9810874704491725, | |
| "grad_norm": 1.359375, | |
| "learning_rate": 5.2039429303079294e-06, | |
| "loss": 0.1892, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 1.9936958234830575, | |
| "grad_norm": 1.1484375, | |
| "learning_rate": 5.152975766325631e-06, | |
| "loss": 0.1953, | |
| "step": 159 | |
| }, | |
| { | |
| "epoch": 2.0063041765169425, | |
| "grad_norm": 1.1015625, | |
| "learning_rate": 5.101992686957028e-06, | |
| "loss": 0.1973, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.0063041765169425, | |
| "eval_loss": 0.2518707811832428, | |
| "eval_runtime": 133.1371, | |
| "eval_samples_per_second": 1.48, | |
| "eval_steps_per_second": 1.48, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.005516154452325, | |
| "grad_norm": 1.3125, | |
| "learning_rate": 5.050998996410899e-06, | |
| "loss": 0.1944, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 2.0181245074862098, | |
| "grad_norm": 1.7421875, | |
| "learning_rate": 5e-06, | |
| "loss": 0.1792, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 2.0307328605200947, | |
| "grad_norm": 1.046875, | |
| "learning_rate": 4.949001003589102e-06, | |
| "loss": 0.1645, | |
| "step": 163 | |
| }, | |
| { | |
| "epoch": 2.0433412135539797, | |
| "grad_norm": 1.109375, | |
| "learning_rate": 4.898007313042975e-06, | |
| "loss": 0.1714, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 2.0559495665878647, | |
| "grad_norm": 1.1640625, | |
| "learning_rate": 4.84702423367437e-06, | |
| "loss": 0.1885, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 2.0685579196217496, | |
| "grad_norm": 1.0625, | |
| "learning_rate": 4.796057069692073e-06, | |
| "loss": 0.1682, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 2.0811662726556346, | |
| "grad_norm": 1.15625, | |
| "learning_rate": 4.74511112364905e-06, | |
| "loss": 0.15, | |
| "step": 167 | |
| }, | |
| { | |
| "epoch": 2.093774625689519, | |
| "grad_norm": 1.078125, | |
| "learning_rate": 4.694191695890788e-06, | |
| "loss": 0.1426, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 2.106382978723404, | |
| "grad_norm": 1.09375, | |
| "learning_rate": 4.643304084003839e-06, | |
| "loss": 0.1286, | |
| "step": 169 | |
| }, | |
| { | |
| "epoch": 2.118991331757289, | |
| "grad_norm": 1.09375, | |
| "learning_rate": 4.592453582264684e-06, | |
| "loss": 0.1593, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.131599684791174, | |
| "grad_norm": 1.1171875, | |
| "learning_rate": 4.541645481088914e-06, | |
| "loss": 0.1571, | |
| "step": 171 | |
| }, | |
| { | |
| "epoch": 2.144208037825059, | |
| "grad_norm": 1.9921875, | |
| "learning_rate": 4.4908850664808245e-06, | |
| "loss": 0.1735, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 2.156816390858944, | |
| "grad_norm": 1.21875, | |
| "learning_rate": 4.4401776194834615e-06, | |
| "loss": 0.1501, | |
| "step": 173 | |
| }, | |
| { | |
| "epoch": 2.169424743892829, | |
| "grad_norm": 1.1796875, | |
| "learning_rate": 4.389528415629201e-06, | |
| "loss": 0.1635, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 2.182033096926714, | |
| "grad_norm": 1.15625, | |
| "learning_rate": 4.33894272439088e-06, | |
| "loss": 0.1733, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 2.194641449960599, | |
| "grad_norm": 1.5390625, | |
| "learning_rate": 4.2884258086335755e-06, | |
| "loss": 0.1733, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 2.207249802994484, | |
| "grad_norm": 1.203125, | |
| "learning_rate": 4.23798292406705e-06, | |
| "loss": 0.1591, | |
| "step": 177 | |
| }, | |
| { | |
| "epoch": 2.219858156028369, | |
| "grad_norm": 1.34375, | |
| "learning_rate": 4.187619318698971e-06, | |
| "loss": 0.1673, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 2.2324665090622537, | |
| "grad_norm": 1.2578125, | |
| "learning_rate": 4.137340232288908e-06, | |
| "loss": 0.162, | |
| "step": 179 | |
| }, | |
| { | |
| "epoch": 2.2450748620961387, | |
| "grad_norm": 1.203125, | |
| "learning_rate": 4.087150895803192e-06, | |
| "loss": 0.1563, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.2450748620961387, | |
| "eval_loss": 0.25570106506347656, | |
| "eval_runtime": 133.1822, | |
| "eval_samples_per_second": 1.479, | |
| "eval_steps_per_second": 1.479, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.2576832151300237, | |
| "grad_norm": 1.1875, | |
| "learning_rate": 4.0370565308706986e-06, | |
| "loss": 0.1609, | |
| "step": 181 | |
| }, | |
| { | |
| "epoch": 2.2702915681639086, | |
| "grad_norm": 2.515625, | |
| "learning_rate": 3.987062349239596e-06, | |
| "loss": 0.1492, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 2.2828999211977936, | |
| "grad_norm": 1.4921875, | |
| "learning_rate": 3.937173552235117e-06, | |
| "loss": 0.1638, | |
| "step": 183 | |
| }, | |
| { | |
| "epoch": 2.2955082742316786, | |
| "grad_norm": 1.21875, | |
| "learning_rate": 3.887395330218429e-06, | |
| "loss": 0.1519, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 2.3081166272655635, | |
| "grad_norm": 1.375, | |
| "learning_rate": 3.837732862046627e-06, | |
| "loss": 0.1965, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 2.3207249802994485, | |
| "grad_norm": 1.7734375, | |
| "learning_rate": 3.7881913145339387e-06, | |
| "loss": 0.1789, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 2.3333333333333335, | |
| "grad_norm": 1.1796875, | |
| "learning_rate": 3.738775841914175e-06, | |
| "loss": 0.1183, | |
| "step": 187 | |
| }, | |
| { | |
| "epoch": 2.3459416863672184, | |
| "grad_norm": 1.3671875, | |
| "learning_rate": 3.689491585304491e-06, | |
| "loss": 0.1702, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 2.3585500394011034, | |
| "grad_norm": 2.03125, | |
| "learning_rate": 3.640343672170503e-06, | |
| "loss": 0.1411, | |
| "step": 189 | |
| }, | |
| { | |
| "epoch": 2.371158392434988, | |
| "grad_norm": 1.3359375, | |
| "learning_rate": 3.5913372157928515e-06, | |
| "loss": 0.1556, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.383766745468873, | |
| "grad_norm": 2.25, | |
| "learning_rate": 3.5424773147352085e-06, | |
| "loss": 0.1785, | |
| "step": 191 | |
| }, | |
| { | |
| "epoch": 2.396375098502758, | |
| "grad_norm": 1.875, | |
| "learning_rate": 3.4937690523138302e-06, | |
| "loss": 0.172, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 2.408983451536643, | |
| "grad_norm": 1.5234375, | |
| "learning_rate": 3.4452174960687033e-06, | |
| "loss": 0.1099, | |
| "step": 193 | |
| }, | |
| { | |
| "epoch": 2.421591804570528, | |
| "grad_norm": 1.296875, | |
| "learning_rate": 3.3968276972363224e-06, | |
| "loss": 0.1552, | |
| "step": 194 | |
| }, | |
| { | |
| "epoch": 2.4342001576044128, | |
| "grad_norm": 1.4140625, | |
| "learning_rate": 3.3486046902241663e-06, | |
| "loss": 0.175, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 2.4468085106382977, | |
| "grad_norm": 1.59375, | |
| "learning_rate": 3.3005534920869175e-06, | |
| "loss": 0.1549, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 2.4594168636721827, | |
| "grad_norm": 1.328125, | |
| "learning_rate": 3.252679102004509e-06, | |
| "loss": 0.1361, | |
| "step": 197 | |
| }, | |
| { | |
| "epoch": 2.4720252167060677, | |
| "grad_norm": 1.453125, | |
| "learning_rate": 3.204986500762006e-06, | |
| "loss": 0.1414, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 2.4846335697399526, | |
| "grad_norm": 1.328125, | |
| "learning_rate": 3.1574806502314206e-06, | |
| "loss": 0.1546, | |
| "step": 199 | |
| }, | |
| { | |
| "epoch": 2.4972419227738376, | |
| "grad_norm": 1.2734375, | |
| "learning_rate": 3.110166492855468e-06, | |
| "loss": 0.1768, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.4972419227738376, | |
| "eval_loss": 0.2556803226470947, | |
| "eval_runtime": 135.2961, | |
| "eval_samples_per_second": 1.456, | |
| "eval_steps_per_second": 1.456, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.5098502758077226, | |
| "grad_norm": 1.2890625, | |
| "learning_rate": 3.063048951133386e-06, | |
| "loss": 0.1709, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 2.5224586288416075, | |
| "grad_norm": 1.3125, | |
| "learning_rate": 3.016132927108787e-06, | |
| "loss": 0.1599, | |
| "step": 202 | |
| }, | |
| { | |
| "epoch": 2.5350669818754925, | |
| "grad_norm": 1.7109375, | |
| "learning_rate": 2.9694233018596665e-06, | |
| "loss": 0.207, | |
| "step": 203 | |
| }, | |
| { | |
| "epoch": 2.5476753349093775, | |
| "grad_norm": 1.59375, | |
| "learning_rate": 2.9229249349905686e-06, | |
| "loss": 0.1585, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 2.5602836879432624, | |
| "grad_norm": 2.625, | |
| "learning_rate": 2.8766426641270197e-06, | |
| "loss": 0.1406, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 2.5728920409771474, | |
| "grad_norm": 1.2421875, | |
| "learning_rate": 2.83058130441221e-06, | |
| "loss": 0.1579, | |
| "step": 206 | |
| }, | |
| { | |
| "epoch": 2.5855003940110324, | |
| "grad_norm": 1.1484375, | |
| "learning_rate": 2.7847456480060476e-06, | |
| "loss": 0.1567, | |
| "step": 207 | |
| }, | |
| { | |
| "epoch": 2.5981087470449173, | |
| "grad_norm": 1.671875, | |
| "learning_rate": 2.7391404635865725e-06, | |
| "loss": 0.2084, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 2.6107171000788023, | |
| "grad_norm": 2.234375, | |
| "learning_rate": 2.6937704958538483e-06, | |
| "loss": 0.1553, | |
| "step": 209 | |
| }, | |
| { | |
| "epoch": 2.6233254531126873, | |
| "grad_norm": 1.609375, | |
| "learning_rate": 2.648640465036316e-06, | |
| "loss": 0.1707, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.6359338061465722, | |
| "grad_norm": 1.2890625, | |
| "learning_rate": 2.603755066399718e-06, | |
| "loss": 0.1355, | |
| "step": 211 | |
| }, | |
| { | |
| "epoch": 2.648542159180457, | |
| "grad_norm": 2.1875, | |
| "learning_rate": 2.559118969758595e-06, | |
| "loss": 0.1576, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 2.661150512214342, | |
| "grad_norm": 1.328125, | |
| "learning_rate": 2.514736818990463e-06, | |
| "loss": 0.1648, | |
| "step": 213 | |
| }, | |
| { | |
| "epoch": 2.673758865248227, | |
| "grad_norm": 1.2734375, | |
| "learning_rate": 2.470613231552661e-06, | |
| "loss": 0.1822, | |
| "step": 214 | |
| }, | |
| { | |
| "epoch": 2.686367218282112, | |
| "grad_norm": 1.2578125, | |
| "learning_rate": 2.4267527980019523e-06, | |
| "loss": 0.1897, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 2.698975571315997, | |
| "grad_norm": 1.2265625, | |
| "learning_rate": 2.383160081516941e-06, | |
| "loss": 0.1548, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 2.711583924349882, | |
| "grad_norm": 1.171875, | |
| "learning_rate": 2.339839617423318e-06, | |
| "loss": 0.1585, | |
| "step": 217 | |
| }, | |
| { | |
| "epoch": 2.724192277383767, | |
| "grad_norm": 1.25, | |
| "learning_rate": 2.296795912722014e-06, | |
| "loss": 0.1975, | |
| "step": 218 | |
| }, | |
| { | |
| "epoch": 2.7368006304176515, | |
| "grad_norm": 1.3203125, | |
| "learning_rate": 2.254033445620293e-06, | |
| "loss": 0.1739, | |
| "step": 219 | |
| }, | |
| { | |
| "epoch": 2.7494089834515365, | |
| "grad_norm": 1.3515625, | |
| "learning_rate": 2.211556665065854e-06, | |
| "loss": 0.1416, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.7494089834515365, | |
| "eval_loss": 0.25502967834472656, | |
| "eval_runtime": 133.2494, | |
| "eval_samples_per_second": 1.478, | |
| "eval_steps_per_second": 1.478, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.7620173364854215, | |
| "grad_norm": 1.1953125, | |
| "learning_rate": 2.169369990283963e-06, | |
| "loss": 0.1772, | |
| "step": 221 | |
| }, | |
| { | |
| "epoch": 2.7746256895193064, | |
| "grad_norm": 1.5390625, | |
| "learning_rate": 2.1274778103176854e-06, | |
| "loss": 0.1667, | |
| "step": 222 | |
| }, | |
| { | |
| "epoch": 2.7872340425531914, | |
| "grad_norm": 1.359375, | |
| "learning_rate": 2.08588448357125e-06, | |
| "loss": 0.1634, | |
| "step": 223 | |
| }, | |
| { | |
| "epoch": 2.7998423955870764, | |
| "grad_norm": 1.7265625, | |
| "learning_rate": 2.0445943373566178e-06, | |
| "loss": 0.1729, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 2.8124507486209613, | |
| "grad_norm": 1.1640625, | |
| "learning_rate": 2.0036116674432653e-06, | |
| "loss": 0.1522, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 2.8250591016548463, | |
| "grad_norm": 1.5625, | |
| "learning_rate": 1.962940737611264e-06, | |
| "loss": 0.1839, | |
| "step": 226 | |
| }, | |
| { | |
| "epoch": 2.8376674546887313, | |
| "grad_norm": 1.5390625, | |
| "learning_rate": 1.922585779207674e-06, | |
| "loss": 0.1673, | |
| "step": 227 | |
| }, | |
| { | |
| "epoch": 2.8502758077226162, | |
| "grad_norm": 1.3359375, | |
| "learning_rate": 1.8825509907063328e-06, | |
| "loss": 0.1527, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 2.862884160756501, | |
| "grad_norm": 1.4296875, | |
| "learning_rate": 1.8428405372710446e-06, | |
| "loss": 0.1735, | |
| "step": 229 | |
| }, | |
| { | |
| "epoch": 2.875492513790386, | |
| "grad_norm": 2.03125, | |
| "learning_rate": 1.8034585503222441e-06, | |
| "loss": 0.1334, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.888100866824271, | |
| "grad_norm": 1.1640625, | |
| "learning_rate": 1.7644091271071645e-06, | |
| "loss": 0.1794, | |
| "step": 231 | |
| }, | |
| { | |
| "epoch": 2.900709219858156, | |
| "grad_norm": 1.3359375, | |
| "learning_rate": 1.7256963302735752e-06, | |
| "loss": 0.1529, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 2.913317572892041, | |
| "grad_norm": 1.2421875, | |
| "learning_rate": 1.687324187447102e-06, | |
| "loss": 0.147, | |
| "step": 233 | |
| }, | |
| { | |
| "epoch": 2.925925925925926, | |
| "grad_norm": 1.21875, | |
| "learning_rate": 1.6492966908122033e-06, | |
| "loss": 0.1458, | |
| "step": 234 | |
| }, | |
| { | |
| "epoch": 2.938534278959811, | |
| "grad_norm": 1.265625, | |
| "learning_rate": 1.611617796696821e-06, | |
| "loss": 0.1467, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 2.951142631993696, | |
| "grad_norm": 1.3828125, | |
| "learning_rate": 1.5742914251607794e-06, | |
| "loss": 0.1492, | |
| "step": 236 | |
| }, | |
| { | |
| "epoch": 2.9637509850275805, | |
| "grad_norm": 1.234375, | |
| "learning_rate": 1.5373214595879416e-06, | |
| "loss": 0.1675, | |
| "step": 237 | |
| }, | |
| { | |
| "epoch": 2.9763593380614655, | |
| "grad_norm": 1.34375, | |
| "learning_rate": 1.500711746282192e-06, | |
| "loss": 0.195, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 2.9889676910953504, | |
| "grad_norm": 1.4453125, | |
| "learning_rate": 1.4644660940672628e-06, | |
| "loss": 0.1617, | |
| "step": 239 | |
| }, | |
| { | |
| "epoch": 3.0015760441292354, | |
| "grad_norm": 1.203125, | |
| "learning_rate": 1.4285882738904822e-06, | |
| "loss": 0.1478, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 3.0015760441292354, | |
| "eval_loss": 0.25495657324790955, | |
| "eval_runtime": 133.3289, | |
| "eval_samples_per_second": 1.478, | |
| "eval_steps_per_second": 1.478, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 3.003940110323089, | |
| "grad_norm": 1.1875, | |
| "learning_rate": 1.3930820184304423e-06, | |
| "loss": 0.1623, | |
| "step": 241 | |
| }, | |
| { | |
| "epoch": 3.016548463356974, | |
| "grad_norm": 1.203125, | |
| "learning_rate": 1.357951021708655e-06, | |
| "loss": 0.1734, | |
| "step": 242 | |
| }, | |
| { | |
| "epoch": 3.029156816390859, | |
| "grad_norm": 1.2109375, | |
| "learning_rate": 1.323198938705238e-06, | |
| "loss": 0.1626, | |
| "step": 243 | |
| }, | |
| { | |
| "epoch": 3.041765169424744, | |
| "grad_norm": 1.546875, | |
| "learning_rate": 1.2888293849786503e-06, | |
| "loss": 0.1478, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.054373522458629, | |
| "grad_norm": 1.109375, | |
| "learning_rate": 1.2548459362895377e-06, | |
| "loss": 0.1453, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 3.0669818754925138, | |
| "grad_norm": 1.1640625, | |
| "learning_rate": 1.2212521282287093e-06, | |
| "loss": 0.1637, | |
| "step": 246 | |
| }, | |
| { | |
| "epoch": 3.0795902285263987, | |
| "grad_norm": 1.125, | |
| "learning_rate": 1.188051455849309e-06, | |
| "loss": 0.1336, | |
| "step": 247 | |
| }, | |
| { | |
| "epoch": 3.0921985815602837, | |
| "grad_norm": 1.375, | |
| "learning_rate": 1.1552473733031893e-06, | |
| "loss": 0.13, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 3.1048069345941687, | |
| "grad_norm": 1.203125, | |
| "learning_rate": 1.1228432934815487e-06, | |
| "loss": 0.1794, | |
| "step": 249 | |
| }, | |
| { | |
| "epoch": 3.1174152876280536, | |
| "grad_norm": 1.2109375, | |
| "learning_rate": 1.0908425876598512e-06, | |
| "loss": 0.1428, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 3.1300236406619386, | |
| "grad_norm": 1.2890625, | |
| "learning_rate": 1.0592485851470973e-06, | |
| "loss": 0.1375, | |
| "step": 251 | |
| }, | |
| { | |
| "epoch": 3.1426319936958236, | |
| "grad_norm": 1.3671875, | |
| "learning_rate": 1.0280645729394368e-06, | |
| "loss": 0.1623, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 3.1552403467297085, | |
| "grad_norm": 1.21875, | |
| "learning_rate": 9.972937953781985e-07, | |
| "loss": 0.1258, | |
| "step": 253 | |
| }, | |
| { | |
| "epoch": 3.1678486997635935, | |
| "grad_norm": 1.1796875, | |
| "learning_rate": 9.6693945381235e-07, | |
| "loss": 0.1696, | |
| "step": 254 | |
| }, | |
| { | |
| "epoch": 3.1804570527974785, | |
| "grad_norm": 1.1328125, | |
| "learning_rate": 9.370047062654386e-07, | |
| "loss": 0.1645, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 3.1930654058313634, | |
| "grad_norm": 1.4765625, | |
| "learning_rate": 9.074926671070322e-07, | |
| "loss": 0.1383, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 3.2056737588652484, | |
| "grad_norm": 1.4453125, | |
| "learning_rate": 8.784064067287057e-07, | |
| "loss": 0.1656, | |
| "step": 257 | |
| }, | |
| { | |
| "epoch": 3.2182821118991334, | |
| "grad_norm": 1.125, | |
| "learning_rate": 8.497489512245971e-07, | |
| "loss": 0.1726, | |
| "step": 258 | |
| }, | |
| { | |
| "epoch": 3.230890464933018, | |
| "grad_norm": 1.171875, | |
| "learning_rate": 8.215232820765851e-07, | |
| "loss": 0.13, | |
| "step": 259 | |
| }, | |
| { | |
| "epoch": 3.243498817966903, | |
| "grad_norm": 1.2578125, | |
| "learning_rate": 7.937323358440935e-07, | |
| "loss": 0.1554, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 3.243498817966903, | |
| "eval_loss": 0.2552575469017029, | |
| "eval_runtime": 133.4201, | |
| "eval_samples_per_second": 1.477, | |
| "eval_steps_per_second": 1.477, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 3.256107171000788, | |
| "grad_norm": 1.2265625, | |
| "learning_rate": 7.663790038585794e-07, | |
| "loss": 0.1601, | |
| "step": 261 | |
| }, | |
| { | |
| "epoch": 3.268715524034673, | |
| "grad_norm": 1.28125, | |
| "learning_rate": 7.394661319227175e-07, | |
| "loss": 0.141, | |
| "step": 262 | |
| }, | |
| { | |
| "epoch": 3.2813238770685578, | |
| "grad_norm": 1.3515625, | |
| "learning_rate": 7.129965200143335e-07, | |
| "loss": 0.1674, | |
| "step": 263 | |
| }, | |
| { | |
| "epoch": 3.2939322301024427, | |
| "grad_norm": 1.265625, | |
| "learning_rate": 6.86972921995096e-07, | |
| "loss": 0.1503, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 3.3065405831363277, | |
| "grad_norm": 1.578125, | |
| "learning_rate": 6.613980453240065e-07, | |
| "loss": 0.1636, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 3.3191489361702127, | |
| "grad_norm": 1.171875, | |
| "learning_rate": 6.36274550775719e-07, | |
| "loss": 0.1622, | |
| "step": 266 | |
| }, | |
| { | |
| "epoch": 3.3317572892040976, | |
| "grad_norm": 1.1328125, | |
| "learning_rate": 6.116050521637218e-07, | |
| "loss": 0.1739, | |
| "step": 267 | |
| }, | |
| { | |
| "epoch": 3.3443656422379826, | |
| "grad_norm": 1.25, | |
| "learning_rate": 5.873921160683943e-07, | |
| "loss": 0.1558, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 3.3569739952718676, | |
| "grad_norm": 1.3828125, | |
| "learning_rate": 5.636382615699842e-07, | |
| "loss": 0.1317, | |
| "step": 269 | |
| }, | |
| { | |
| "epoch": 3.3695823483057525, | |
| "grad_norm": 1.2578125, | |
| "learning_rate": 5.403459599865307e-07, | |
| "loss": 0.1626, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 3.3821907013396375, | |
| "grad_norm": 1.453125, | |
| "learning_rate": 5.175176346167465e-07, | |
| "loss": 0.1652, | |
| "step": 271 | |
| }, | |
| { | |
| "epoch": 3.3947990543735225, | |
| "grad_norm": 1.3046875, | |
| "learning_rate": 4.951556604879049e-07, | |
| "loss": 0.1606, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 3.4074074074074074, | |
| "grad_norm": 1.296875, | |
| "learning_rate": 4.732623641087403e-07, | |
| "loss": 0.1689, | |
| "step": 273 | |
| }, | |
| { | |
| "epoch": 3.4200157604412924, | |
| "grad_norm": 1.2578125, | |
| "learning_rate": 4.5184002322740784e-07, | |
| "loss": 0.1437, | |
| "step": 274 | |
| }, | |
| { | |
| "epoch": 3.4326241134751774, | |
| "grad_norm": 1.109375, | |
| "learning_rate": 4.3089086659450774e-07, | |
| "loss": 0.1587, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 3.4452324665090623, | |
| "grad_norm": 1.2734375, | |
| "learning_rate": 4.1041707373120354e-07, | |
| "loss": 0.1378, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 3.4578408195429473, | |
| "grad_norm": 1.171875, | |
| "learning_rate": 3.9042077470247574e-07, | |
| "loss": 0.1696, | |
| "step": 277 | |
| }, | |
| { | |
| "epoch": 3.4704491725768323, | |
| "grad_norm": 1.8515625, | |
| "learning_rate": 3.709040498955102e-07, | |
| "loss": 0.1644, | |
| "step": 278 | |
| }, | |
| { | |
| "epoch": 3.4830575256107172, | |
| "grad_norm": 1.6875, | |
| "learning_rate": 3.518689298032524e-07, | |
| "loss": 0.163, | |
| "step": 279 | |
| }, | |
| { | |
| "epoch": 3.495665878644602, | |
| "grad_norm": 1.1171875, | |
| "learning_rate": 3.333173948131663e-07, | |
| "loss": 0.1325, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 3.495665878644602, | |
| "eval_loss": 0.2554157078266144, | |
| "eval_runtime": 135.0407, | |
| "eval_samples_per_second": 1.459, | |
| "eval_steps_per_second": 1.459, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 3.508274231678487, | |
| "grad_norm": 1.2421875, | |
| "learning_rate": 3.1525137500119207e-07, | |
| "loss": 0.1692, | |
| "step": 281 | |
| }, | |
| { | |
| "epoch": 3.520882584712372, | |
| "grad_norm": 1.359375, | |
| "learning_rate": 2.9767274993094285e-07, | |
| "loss": 0.1734, | |
| "step": 282 | |
| }, | |
| { | |
| "epoch": 3.533490937746257, | |
| "grad_norm": 1.5859375, | |
| "learning_rate": 2.8058334845816214e-07, | |
| "loss": 0.1789, | |
| "step": 283 | |
| }, | |
| { | |
| "epoch": 3.546099290780142, | |
| "grad_norm": 1.171875, | |
| "learning_rate": 2.6398494854045055e-07, | |
| "loss": 0.1481, | |
| "step": 284 | |
| }, | |
| { | |
| "epoch": 3.558707643814027, | |
| "grad_norm": 1.3203125, | |
| "learning_rate": 2.478792770522842e-07, | |
| "loss": 0.1596, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 3.571315996847912, | |
| "grad_norm": 1.3671875, | |
| "learning_rate": 2.32268009605362e-07, | |
| "loss": 0.1752, | |
| "step": 286 | |
| }, | |
| { | |
| "epoch": 3.5839243498817965, | |
| "grad_norm": 1.34375, | |
| "learning_rate": 2.171527703742715e-07, | |
| "loss": 0.1298, | |
| "step": 287 | |
| }, | |
| { | |
| "epoch": 3.5965327029156815, | |
| "grad_norm": 1.296875, | |
| "learning_rate": 2.0253513192751374e-07, | |
| "loss": 0.1669, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 3.6091410559495665, | |
| "grad_norm": 1.546875, | |
| "learning_rate": 1.884166150638933e-07, | |
| "loss": 0.1632, | |
| "step": 289 | |
| }, | |
| { | |
| "epoch": 3.6217494089834514, | |
| "grad_norm": 1.1015625, | |
| "learning_rate": 1.7479868865430072e-07, | |
| "loss": 0.1456, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 3.6343577620173364, | |
| "grad_norm": 1.46875, | |
| "learning_rate": 1.6168276948889007e-07, | |
| "loss": 0.1471, | |
| "step": 291 | |
| }, | |
| { | |
| "epoch": 3.6469661150512214, | |
| "grad_norm": 1.21875, | |
| "learning_rate": 1.4907022212967803e-07, | |
| "loss": 0.1613, | |
| "step": 292 | |
| }, | |
| { | |
| "epoch": 3.6595744680851063, | |
| "grad_norm": 1.328125, | |
| "learning_rate": 1.3696235876857812e-07, | |
| "loss": 0.1641, | |
| "step": 293 | |
| }, | |
| { | |
| "epoch": 3.6721828211189913, | |
| "grad_norm": 1.5234375, | |
| "learning_rate": 1.253604390908819e-07, | |
| "loss": 0.1615, | |
| "step": 294 | |
| }, | |
| { | |
| "epoch": 3.6847911741528763, | |
| "grad_norm": 1.40625, | |
| "learning_rate": 1.1426567014420297e-07, | |
| "loss": 0.1428, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 3.6973995271867612, | |
| "grad_norm": 1.1953125, | |
| "learning_rate": 1.0367920621289496e-07, | |
| "loss": 0.1876, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 3.710007880220646, | |
| "grad_norm": 1.3671875, | |
| "learning_rate": 9.360214869796492e-08, | |
| "loss": 0.149, | |
| "step": 297 | |
| }, | |
| { | |
| "epoch": 3.722616233254531, | |
| "grad_norm": 1.34375, | |
| "learning_rate": 8.403554600248498e-08, | |
| "loss": 0.1671, | |
| "step": 298 | |
| }, | |
| { | |
| "epoch": 3.735224586288416, | |
| "grad_norm": 1.0703125, | |
| "learning_rate": 7.498039342251573e-08, | |
| "loss": 0.1411, | |
| "step": 299 | |
| }, | |
| { | |
| "epoch": 3.747832939322301, | |
| "grad_norm": 1.390625, | |
| "learning_rate": 6.643763304355566e-08, | |
| "loss": 0.1366, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.747832939322301, | |
| "eval_loss": 0.2554207742214203, | |
| "eval_runtime": 135.9935, | |
| "eval_samples_per_second": 1.449, | |
| "eval_steps_per_second": 1.449, | |
| "step": 300 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 316, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 20, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.7706338758950912e+18, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |