| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0016680567139282, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 2e-05, | |
| "loss": 1.7915, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 4e-05, | |
| "loss": 1.7449, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 6e-05, | |
| "loss": 1.7538, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 8e-05, | |
| "loss": 1.713, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.0001, | |
| "loss": 1.7313, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_loss": 1.6295605897903442, | |
| "eval_runtime": 6.0457, | |
| "eval_samples_per_second": 0.331, | |
| "eval_steps_per_second": 0.331, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.00012, | |
| "loss": 1.8184, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.00014, | |
| "loss": 1.7586, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.00016, | |
| "loss": 1.8926, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.00018, | |
| "loss": 1.9176, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0002, | |
| "loss": 1.584, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_loss": 1.606095790863037, | |
| "eval_runtime": 6.0412, | |
| "eval_samples_per_second": 0.331, | |
| "eval_steps_per_second": 0.331, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.00019997370884991842, | |
| "loss": 1.2008, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.00019989484922416502, | |
| "loss": 1.2531, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.00019976346258894503, | |
| "loss": 1.6787, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.00019957961803037326, | |
| "loss": 1.7146, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.00019934341221814739, | |
| "loss": 1.6937, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_loss": 1.5841087102890015, | |
| "eval_runtime": 6.0341, | |
| "eval_samples_per_second": 0.331, | |
| "eval_steps_per_second": 0.331, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.00019905496935471658, | |
| "loss": 1.7531, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.0001987144411099731, | |
| "loss": 1.7094, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.00019832200654150076, | |
| "loss": 1.7333, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.00019787787200042223, | |
| "loss": 1.7736, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0001973822710228951, | |
| "loss": 1.6655, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 1.5755609273910522, | |
| "eval_runtime": 6.0389, | |
| "eval_samples_per_second": 0.331, | |
| "eval_steps_per_second": 0.331, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.0001968354642073129, | |
| "loss": 1.8054, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.00019623773907727682, | |
| "loss": 1.8327, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.00019558940993040885, | |
| "loss": 1.3822, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.00019489081767308698, | |
| "loss": 1.254, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.00019414232964118892, | |
| "loss": 1.7648, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 1.5711314678192139, | |
| "eval_runtime": 6.0254, | |
| "eval_samples_per_second": 0.332, | |
| "eval_steps_per_second": 0.332, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.0001933443394069383, | |
| "loss": 1.641, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.00019249726657195532, | |
| "loss": 1.7777, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.00019160155654662076, | |
| "loss": 1.7038, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.0001906576803158686, | |
| "loss": 1.6179, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.0001896661341915318, | |
| "loss": 1.7813, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_loss": 1.5650010108947754, | |
| "eval_runtime": 6.0228, | |
| "eval_samples_per_second": 0.332, | |
| "eval_steps_per_second": 0.332, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.00018862743955136966, | |
| "loss": 1.661, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 0.00018754214256491562, | |
| "loss": 1.7948, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.00018641081390628877, | |
| "loss": 1.92, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.00018523404845412027, | |
| "loss": 1.6941, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.0001840124649787524, | |
| "loss": 1.3461, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "eval_loss": 1.5624847412109375, | |
| "eval_runtime": 6.0325, | |
| "eval_samples_per_second": 0.332, | |
| "eval_steps_per_second": 0.332, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.0001827467058168748, | |
| "loss": 0.8176, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 0.00018143743653376942, | |
| "loss": 1.7262, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.00018008534557334064, | |
| "loss": 1.7333, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.00017869114389611575, | |
| "loss": 1.5991, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0001772555646054055, | |
| "loss": 1.7267, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 1.5579214096069336, | |
| "eval_runtime": 6.0388, | |
| "eval_samples_per_second": 0.331, | |
| "eval_steps_per_second": 0.331, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 0.00017577936256182167, | |
| "loss": 1.6694, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 0.0001742633139863538, | |
| "loss": 1.8201, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 0.0001727082160522145, | |
| "loss": 1.7913, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 0.00017111488646566727, | |
| "loss": 1.825, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 0.00016948416303605795, | |
| "loss": 1.7778, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "eval_loss": 1.555617094039917, | |
| "eval_runtime": 6.0421, | |
| "eval_samples_per_second": 0.331, | |
| "eval_steps_per_second": 0.331, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 0.00016781690323527511, | |
| "loss": 1.6311, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 0.0001661139837468717, | |
| "loss": 1.1499, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 0.00016437630000508464, | |
| "loss": 1.0455, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 0.00016260476572399496, | |
| "loss": 1.7178, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.00016080031241707578, | |
| "loss": 1.4832, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.554579257965088, | |
| "eval_runtime": 6.0239, | |
| "eval_samples_per_second": 0.332, | |
| "eval_steps_per_second": 0.332, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.00015896388890738127, | |
| "loss": 1.6801, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 0.0001570964608286336, | |
| "loss": 1.6462, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 0.00015519901011747044, | |
| "loss": 1.7264, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 0.0001532725344971202, | |
| "loss": 1.63, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 0.00015131804695277612, | |
| "loss": 1.7584, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "eval_loss": 1.5519572496414185, | |
| "eval_runtime": 6.0181, | |
| "eval_samples_per_second": 0.332, | |
| "eval_steps_per_second": 0.332, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 0.0001493365751989454, | |
| "loss": 1.7956, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 0.00014732916113905335, | |
| "loss": 1.6528, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 0.00014529686031758643, | |
| "loss": 1.7006, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 0.00014324074136506284, | |
| "loss": 1.8171, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.0001411618854361218, | |
| "loss": 1.4825, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "eval_loss": 1.5512959957122803, | |
| "eval_runtime": 6.0143, | |
| "eval_samples_per_second": 0.333, | |
| "eval_steps_per_second": 0.333, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 0.00013906138564102793, | |
| "loss": 0.9645, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 0.0001369403464708884, | |
| "loss": 1.2708, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 0.0001347998832168862, | |
| "loss": 1.7432, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 0.00013264112138383445, | |
| "loss": 1.6986, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 0.00013046519609836, | |
| "loss": 1.7969, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "eval_loss": 1.5507731437683105, | |
| "eval_runtime": 6.0363, | |
| "eval_samples_per_second": 0.331, | |
| "eval_steps_per_second": 0.331, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 0.00012827325151202782, | |
| "loss": 1.5769, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 0.00012606644019971968, | |
| "loss": 1.7011, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 0.00012384592255358385, | |
| "loss": 1.6525, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.00012161286617287419, | |
| "loss": 1.6834, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 0.00011936844524999966, | |
| "loss": 1.7182, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "eval_loss": 1.549034595489502, | |
| "eval_runtime": 6.0315, | |
| "eval_samples_per_second": 0.332, | |
| "eval_steps_per_second": 0.332, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 0.00011711383995310681, | |
| "loss": 1.7941, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 0.00011485023580552039, | |
| "loss": 1.6289, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 0.00011257882306236775, | |
| "loss": 1.1176, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 0.00011030079608471544, | |
| "loss": 0.8403, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 0.00010801735271154669, | |
| "loss": 1.724, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "eval_loss": 1.5494024753570557, | |
| "eval_runtime": 6.0321, | |
| "eval_samples_per_second": 0.332, | |
| "eval_steps_per_second": 0.332, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 0.00010572969362990998, | |
| "loss": 1.6746, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 0.00010343902174357039, | |
| "loss": 1.7152, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 0.0001011465415404949, | |
| "loss": 1.7054, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 9.88534584595051e-05, | |
| "loss": 1.7333, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 9.656097825642961e-05, | |
| "loss": 1.7083, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": 1.5476174354553223, | |
| "eval_runtime": 6.0289, | |
| "eval_samples_per_second": 0.332, | |
| "eval_steps_per_second": 0.332, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 9.427030637009003e-05, | |
| "loss": 1.8771, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 9.198264728845332e-05, | |
| "loss": 1.8446, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 8.969920391528458e-05, | |
| "loss": 1.86, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 8.742117693763227e-05, | |
| "loss": 1.7291, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 8.514976419447964e-05, | |
| "loss": 1.2195, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "eval_loss": 1.5478819608688354, | |
| "eval_runtime": 6.0353, | |
| "eval_samples_per_second": 0.331, | |
| "eval_steps_per_second": 0.331, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 8.28861600468932e-05, | |
| "loss": 0.9047, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 8.063155475000037e-05, | |
| "loss": 1.7297, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 7.838713382712583e-05, | |
| "loss": 1.5731, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 7.615407744641619e-05, | |
| "loss": 1.598, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 7.393355980028039e-05, | |
| "loss": 1.6782, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "eval_loss": 1.5477495193481445, | |
| "eval_runtime": 6.0384, | |
| "eval_samples_per_second": 0.331, | |
| "eval_steps_per_second": 0.331, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 7.172674848797219e-05, | |
| "loss": 1.672, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 6.953480390164e-05, | |
| "loss": 1.6457, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 6.735887861616556e-05, | |
| "loss": 1.7155, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 6.520011678311382e-05, | |
| "loss": 1.7334, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 6.305965352911161e-05, | |
| "loss": 1.7342, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "eval_loss": 1.5465799570083618, | |
| "eval_runtime": 6.0244, | |
| "eval_samples_per_second": 0.332, | |
| "eval_steps_per_second": 0.332, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 6.093861435897208e-05, | |
| "loss": 1.8115, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 5.883811456387821e-05, | |
| "loss": 1.4625, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 5.675925863493721e-05, | |
| "loss": 1.08, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 5.4703139682413586e-05, | |
| "loss": 1.6914, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 5.267083886094668e-05, | |
| "loss": 1.6016, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.547050952911377, | |
| "eval_runtime": 6.0254, | |
| "eval_samples_per_second": 0.332, | |
| "eval_steps_per_second": 0.332, | |
| "step": 100 | |
| } | |
| ], | |
| "max_steps": 147, | |
| "num_train_epochs": 3, | |
| "total_flos": 5.5839227595669504e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |