| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.963768115942029, | |
| "eval_steps": 35, | |
| "global_step": 207, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 43.0, | |
| "learning_rate": 2.0000000000000002e-07, | |
| "loss": 0.7175, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "eval_loss": 0.7699161767959595, | |
| "eval_runtime": 19.565, | |
| "eval_samples_per_second": 200.818, | |
| "eval_steps_per_second": 200.818, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 42.5, | |
| "learning_rate": 4.0000000000000003e-07, | |
| "loss": 0.7271, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 42.5, | |
| "learning_rate": 6.000000000000001e-07, | |
| "loss": 0.7161, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 41.25, | |
| "learning_rate": 8.000000000000001e-07, | |
| "loss": 0.7088, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 42.5, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 0.7022, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 41.0, | |
| "learning_rate": 1.2000000000000002e-06, | |
| "loss": 0.6846, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 38.25, | |
| "learning_rate": 1.4000000000000001e-06, | |
| "loss": 0.629, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 32.5, | |
| "learning_rate": 1.6000000000000001e-06, | |
| "loss": 0.5753, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 28.125, | |
| "learning_rate": 1.8000000000000001e-06, | |
| "loss": 0.5368, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 22.125, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 0.4946, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 19.125, | |
| "learning_rate": 2.2e-06, | |
| "loss": 0.4351, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 15.3125, | |
| "learning_rate": 2.4000000000000003e-06, | |
| "loss": 0.3873, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 12.25, | |
| "learning_rate": 2.6e-06, | |
| "loss": 0.3382, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 10.3125, | |
| "learning_rate": 2.8000000000000003e-06, | |
| "loss": 0.2777, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 8.4375, | |
| "learning_rate": 3e-06, | |
| "loss": 0.241, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 7.46875, | |
| "learning_rate": 3.2000000000000003e-06, | |
| "loss": 0.2278, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 5.65625, | |
| "learning_rate": 3.4000000000000005e-06, | |
| "loss": 0.1924, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 4.65625, | |
| "learning_rate": 3.6000000000000003e-06, | |
| "loss": 0.1635, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 3.546875, | |
| "learning_rate": 3.8000000000000005e-06, | |
| "loss": 0.1549, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 2.5, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.1374, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 2.875, | |
| "learning_rate": 4.2000000000000004e-06, | |
| "loss": 0.12, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 2.59375, | |
| "learning_rate": 4.4e-06, | |
| "loss": 0.118, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 2.453125, | |
| "learning_rate": 4.600000000000001e-06, | |
| "loss": 0.1095, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 2.171875, | |
| "learning_rate": 4.800000000000001e-06, | |
| "loss": 0.0964, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 1.7578125, | |
| "learning_rate": 5e-06, | |
| "loss": 0.0887, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 1.703125, | |
| "learning_rate": 5.2e-06, | |
| "loss": 0.077, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "grad_norm": 1.625, | |
| "learning_rate": 5.400000000000001e-06, | |
| "loss": 0.074, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 1.4609375, | |
| "learning_rate": 5.600000000000001e-06, | |
| "loss": 0.0659, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 1.140625, | |
| "learning_rate": 5.8e-06, | |
| "loss": 0.0586, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 2.265625, | |
| "learning_rate": 6e-06, | |
| "loss": 0.0518, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 1.609375, | |
| "learning_rate": 6.200000000000001e-06, | |
| "loss": 0.0566, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 1.453125, | |
| "learning_rate": 6.4000000000000006e-06, | |
| "loss": 0.0475, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 1.296875, | |
| "learning_rate": 6.600000000000001e-06, | |
| "loss": 0.0495, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 1.234375, | |
| "learning_rate": 6.800000000000001e-06, | |
| "loss": 0.0493, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 1.03125, | |
| "learning_rate": 7e-06, | |
| "loss": 0.055, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_loss": 0.03942573070526123, | |
| "eval_runtime": 19.6691, | |
| "eval_samples_per_second": 199.755, | |
| "eval_steps_per_second": 199.755, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 0.94921875, | |
| "learning_rate": 7.2000000000000005e-06, | |
| "loss": 0.0457, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 1.09375, | |
| "learning_rate": 7.4e-06, | |
| "loss": 0.0452, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 0.83203125, | |
| "learning_rate": 7.600000000000001e-06, | |
| "loss": 0.0373, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 0.9296875, | |
| "learning_rate": 7.800000000000002e-06, | |
| "loss": 0.0379, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 1.0625, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.0403, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 0.91015625, | |
| "learning_rate": 8.2e-06, | |
| "loss": 0.0445, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 0.96484375, | |
| "learning_rate": 8.400000000000001e-06, | |
| "loss": 0.0377, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 0.73046875, | |
| "learning_rate": 8.6e-06, | |
| "loss": 0.0367, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.88671875, | |
| "learning_rate": 8.8e-06, | |
| "loss": 0.035, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 0.94921875, | |
| "learning_rate": 9e-06, | |
| "loss": 0.0417, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 0.81640625, | |
| "learning_rate": 9.200000000000002e-06, | |
| "loss": 0.0279, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 0.91796875, | |
| "learning_rate": 9.4e-06, | |
| "loss": 0.0292, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 1.5625, | |
| "learning_rate": 9.600000000000001e-06, | |
| "loss": 0.0345, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 0.8515625, | |
| "learning_rate": 9.800000000000001e-06, | |
| "loss": 0.0348, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.98046875, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0337, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 0.75, | |
| "learning_rate": 1.02e-05, | |
| "loss": 0.0361, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.71875, | |
| "learning_rate": 1.04e-05, | |
| "loss": 0.0363, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 0.7265625, | |
| "learning_rate": 1.0600000000000002e-05, | |
| "loss": 0.0329, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 0.68359375, | |
| "learning_rate": 1.0800000000000002e-05, | |
| "loss": 0.032, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.46875, | |
| "learning_rate": 1.1000000000000001e-05, | |
| "loss": 0.0275, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 0.62890625, | |
| "learning_rate": 1.1200000000000001e-05, | |
| "loss": 0.0304, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "grad_norm": 0.7421875, | |
| "learning_rate": 1.14e-05, | |
| "loss": 0.0325, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 0.51953125, | |
| "learning_rate": 1.16e-05, | |
| "loss": 0.027, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 1.015625, | |
| "learning_rate": 1.18e-05, | |
| "loss": 0.0283, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 0.62890625, | |
| "learning_rate": 1.2e-05, | |
| "loss": 0.0301, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.734375, | |
| "learning_rate": 1.22e-05, | |
| "loss": 0.0334, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 0.6875, | |
| "learning_rate": 1.2400000000000002e-05, | |
| "loss": 0.0312, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 0.58203125, | |
| "learning_rate": 1.2600000000000001e-05, | |
| "loss": 0.0282, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 0.65625, | |
| "learning_rate": 1.2800000000000001e-05, | |
| "loss": 0.0315, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 0.66015625, | |
| "learning_rate": 1.3000000000000001e-05, | |
| "loss": 0.0316, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.55078125, | |
| "learning_rate": 1.3200000000000002e-05, | |
| "loss": 0.0292, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 0.84375, | |
| "learning_rate": 1.3400000000000002e-05, | |
| "loss": 0.0304, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 0.498046875, | |
| "learning_rate": 1.3600000000000002e-05, | |
| "loss": 0.0235, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.6875, | |
| "learning_rate": 1.38e-05, | |
| "loss": 0.0262, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "grad_norm": 0.8046875, | |
| "learning_rate": 1.4e-05, | |
| "loss": 0.03, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "eval_loss": 0.023079946637153625, | |
| "eval_runtime": 19.5963, | |
| "eval_samples_per_second": 200.497, | |
| "eval_steps_per_second": 200.497, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "grad_norm": 0.77734375, | |
| "learning_rate": 1.4200000000000001e-05, | |
| "loss": 0.0226, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "grad_norm": 0.671875, | |
| "learning_rate": 1.4400000000000001e-05, | |
| "loss": 0.0274, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 0.5390625, | |
| "learning_rate": 1.46e-05, | |
| "loss": 0.0241, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 0.6953125, | |
| "learning_rate": 1.48e-05, | |
| "loss": 0.0239, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "grad_norm": 0.5546875, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.0237, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "grad_norm": 0.55859375, | |
| "learning_rate": 1.5200000000000002e-05, | |
| "loss": 0.0231, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 0.55078125, | |
| "learning_rate": 1.54e-05, | |
| "loss": 0.0234, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "grad_norm": 0.5546875, | |
| "learning_rate": 1.5600000000000003e-05, | |
| "loss": 0.0209, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "grad_norm": 0.67578125, | |
| "learning_rate": 1.58e-05, | |
| "loss": 0.0191, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "grad_norm": 0.578125, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.0232, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "grad_norm": 0.640625, | |
| "learning_rate": 1.62e-05, | |
| "loss": 0.0263, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "grad_norm": 0.68359375, | |
| "learning_rate": 1.64e-05, | |
| "loss": 0.0221, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "grad_norm": 0.60546875, | |
| "learning_rate": 1.66e-05, | |
| "loss": 0.0207, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 1.1328125, | |
| "learning_rate": 1.6800000000000002e-05, | |
| "loss": 0.0224, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "grad_norm": 0.6328125, | |
| "learning_rate": 1.7e-05, | |
| "loss": 0.0206, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "grad_norm": 0.8046875, | |
| "learning_rate": 1.72e-05, | |
| "loss": 0.0244, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "grad_norm": 0.7265625, | |
| "learning_rate": 1.7400000000000003e-05, | |
| "loss": 0.0223, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "grad_norm": 0.466796875, | |
| "learning_rate": 1.76e-05, | |
| "loss": 0.02, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "grad_norm": 0.67578125, | |
| "learning_rate": 1.7800000000000002e-05, | |
| "loss": 0.0237, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "grad_norm": 0.69921875, | |
| "learning_rate": 1.8e-05, | |
| "loss": 0.0222, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 0.46484375, | |
| "learning_rate": 1.8200000000000002e-05, | |
| "loss": 0.0225, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "grad_norm": 0.5546875, | |
| "learning_rate": 1.8400000000000003e-05, | |
| "loss": 0.0204, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "grad_norm": 0.51171875, | |
| "learning_rate": 1.86e-05, | |
| "loss": 0.0199, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "grad_norm": 0.41796875, | |
| "learning_rate": 1.88e-05, | |
| "loss": 0.0202, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "grad_norm": 0.443359375, | |
| "learning_rate": 1.9e-05, | |
| "loss": 0.021, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "grad_norm": 0.7421875, | |
| "learning_rate": 1.9200000000000003e-05, | |
| "loss": 0.0219, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "grad_norm": 0.60546875, | |
| "learning_rate": 1.94e-05, | |
| "loss": 0.021, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 0.55078125, | |
| "learning_rate": 1.9600000000000002e-05, | |
| "loss": 0.0207, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "grad_norm": 0.62890625, | |
| "learning_rate": 1.98e-05, | |
| "loss": 0.0181, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "grad_norm": 0.486328125, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0203, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "grad_norm": 0.75390625, | |
| "learning_rate": 1.9995690062269985e-05, | |
| "loss": 0.0194, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "grad_norm": 0.56640625, | |
| "learning_rate": 1.9982763964192586e-05, | |
| "loss": 0.0207, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "grad_norm": 0.482421875, | |
| "learning_rate": 1.996123284790336e-05, | |
| "loss": 0.0204, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "grad_norm": 0.70703125, | |
| "learning_rate": 1.9931115272956405e-05, | |
| "loss": 0.0225, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 0.53515625, | |
| "learning_rate": 1.989243720032624e-05, | |
| "loss": 0.0215, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "eval_loss": 0.020286040380597115, | |
| "eval_runtime": 19.6725, | |
| "eval_samples_per_second": 199.72, | |
| "eval_steps_per_second": 199.72, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "grad_norm": 0.48828125, | |
| "learning_rate": 1.9845231970029774e-05, | |
| "loss": 0.0199, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "grad_norm": 0.5703125, | |
| "learning_rate": 1.978954027238763e-05, | |
| "loss": 0.021, | |
| "step": 107 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "grad_norm": 0.431640625, | |
| "learning_rate": 1.972541011294959e-05, | |
| "loss": 0.0196, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "grad_norm": 0.388671875, | |
| "learning_rate": 1.9652896771114416e-05, | |
| "loss": 0.0193, | |
| "step": 109 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "grad_norm": 0.423828125, | |
| "learning_rate": 1.9572062752479684e-05, | |
| "loss": 0.02, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "grad_norm": 0.53125, | |
| "learning_rate": 1.9482977734962753e-05, | |
| "loss": 0.0219, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "grad_norm": 0.41015625, | |
| "learning_rate": 1.9385718508739263e-05, | |
| "loss": 0.0184, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "grad_norm": 0.50390625, | |
| "learning_rate": 1.9280368910050943e-05, | |
| "loss": 0.019, | |
| "step": 113 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "grad_norm": 0.470703125, | |
| "learning_rate": 1.9167019748939847e-05, | |
| "loss": 0.019, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "grad_norm": 0.515625, | |
| "learning_rate": 1.9045768730971198e-05, | |
| "loss": 0.0237, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "grad_norm": 0.4296875, | |
| "learning_rate": 1.8916720373012425e-05, | |
| "loss": 0.0222, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "grad_norm": 0.42578125, | |
| "learning_rate": 1.8779985913140927e-05, | |
| "loss": 0.0215, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "grad_norm": 0.4375, | |
| "learning_rate": 1.8635683214758213e-05, | |
| "loss": 0.0207, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "grad_norm": 0.431640625, | |
| "learning_rate": 1.8483936664993152e-05, | |
| "loss": 0.0188, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "grad_norm": 0.4921875, | |
| "learning_rate": 1.8324877067481782e-05, | |
| "loss": 0.0199, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "grad_norm": 0.5234375, | |
| "learning_rate": 1.815864152961624e-05, | |
| "loss": 0.0208, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 0.486328125, | |
| "learning_rate": 1.798537334435986e-05, | |
| "loss": 0.0182, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "grad_norm": 0.55859375, | |
| "learning_rate": 1.780522186673046e-05, | |
| "loss": 0.0199, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "grad_norm": 0.35546875, | |
| "learning_rate": 1.7618342385058147e-05, | |
| "loss": 0.0158, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "grad_norm": 0.51171875, | |
| "learning_rate": 1.7424895987128723e-05, | |
| "loss": 0.0206, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "grad_norm": 0.51953125, | |
| "learning_rate": 1.7225049421328024e-05, | |
| "loss": 0.0192, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "grad_norm": 0.380859375, | |
| "learning_rate": 1.7018974952906885e-05, | |
| "loss": 0.0164, | |
| "step": 127 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "grad_norm": 0.41015625, | |
| "learning_rate": 1.680685021549063e-05, | |
| "loss": 0.0186, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "grad_norm": 0.4921875, | |
| "learning_rate": 1.658885805796111e-05, | |
| "loss": 0.0178, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "grad_norm": 0.42578125, | |
| "learning_rate": 1.636518638684325e-05, | |
| "loss": 0.0179, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 0.40234375, | |
| "learning_rate": 1.613602800433194e-05, | |
| "loss": 0.0191, | |
| "step": 131 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "grad_norm": 0.41015625, | |
| "learning_rate": 1.590158044209897e-05, | |
| "loss": 0.0201, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "grad_norm": 0.455078125, | |
| "learning_rate": 1.566204579102317e-05, | |
| "loss": 0.0196, | |
| "step": 133 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.44921875, | |
| "learning_rate": 1.5417630526990613e-05, | |
| "loss": 0.0197, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "grad_norm": 0.37890625, | |
| "learning_rate": 1.5168545332914942e-05, | |
| "loss": 0.0185, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "grad_norm": 0.330078125, | |
| "learning_rate": 1.4915004917131345e-05, | |
| "loss": 0.0162, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "grad_norm": 0.38671875, | |
| "learning_rate": 1.4657227828320637e-05, | |
| "loss": 0.0208, | |
| "step": 137 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "grad_norm": 0.392578125, | |
| "learning_rate": 1.4395436267123017e-05, | |
| "loss": 0.0184, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.388671875, | |
| "learning_rate": 1.4129855894603885e-05, | |
| "loss": 0.0181, | |
| "step": 139 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "grad_norm": 0.5078125, | |
| "learning_rate": 1.3860715637736817e-05, | |
| "loss": 0.0185, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "eval_loss": 0.019323738291859627, | |
| "eval_runtime": 19.6403, | |
| "eval_samples_per_second": 200.047, | |
| "eval_steps_per_second": 200.047, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "grad_norm": 0.3203125, | |
| "learning_rate": 1.358824749207136e-05, | |
| "loss": 0.0139, | |
| "step": 141 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "grad_norm": 0.361328125, | |
| "learning_rate": 1.331268632175576e-05, | |
| "loss": 0.0129, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "grad_norm": 0.26953125, | |
| "learning_rate": 1.3034269657086993e-05, | |
| "loss": 0.0128, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "grad_norm": 0.40234375, | |
| "learning_rate": 1.27532374897626e-05, | |
| "loss": 0.0116, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "grad_norm": 0.357421875, | |
| "learning_rate": 1.2469832066010843e-05, | |
| "loss": 0.011, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 0.3203125, | |
| "learning_rate": 1.2184297677777463e-05, | |
| "loss": 0.0113, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "grad_norm": 0.37890625, | |
| "learning_rate": 1.1896880452149077e-05, | |
| "loss": 0.0113, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "grad_norm": 0.349609375, | |
| "learning_rate": 1.1607828139194683e-05, | |
| "loss": 0.0135, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "grad_norm": 0.298828125, | |
| "learning_rate": 1.1317389898408188e-05, | |
| "loss": 0.0116, | |
| "step": 149 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "grad_norm": 0.27734375, | |
| "learning_rate": 1.1025816083936036e-05, | |
| "loss": 0.0119, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "grad_norm": 0.400390625, | |
| "learning_rate": 1.073335802877504e-05, | |
| "loss": 0.0112, | |
| "step": 151 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "grad_norm": 0.44140625, | |
| "learning_rate": 1.0440267828126478e-05, | |
| "loss": 0.0127, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "grad_norm": 0.271484375, | |
| "learning_rate": 1.0146798122093167e-05, | |
| "loss": 0.011, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 0.30859375, | |
| "learning_rate": 9.853201877906836e-06, | |
| "loss": 0.0111, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "grad_norm": 0.373046875, | |
| "learning_rate": 9.559732171873524e-06, | |
| "loss": 0.0111, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "grad_norm": 0.35546875, | |
| "learning_rate": 9.266641971224963e-06, | |
| "loss": 0.0129, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "grad_norm": 0.3203125, | |
| "learning_rate": 8.974183916063967e-06, | |
| "loss": 0.0114, | |
| "step": 157 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "grad_norm": 0.41015625, | |
| "learning_rate": 8.682610101591813e-06, | |
| "loss": 0.0122, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "grad_norm": 0.400390625, | |
| "learning_rate": 8.39217186080532e-06, | |
| "loss": 0.0108, | |
| "step": 159 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "grad_norm": 0.2451171875, | |
| "learning_rate": 8.103119547850924e-06, | |
| "loss": 0.0099, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "grad_norm": 0.265625, | |
| "learning_rate": 7.815702322222539e-06, | |
| "loss": 0.01, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "grad_norm": 0.38671875, | |
| "learning_rate": 7.530167933989161e-06, | |
| "loss": 0.0122, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "grad_norm": 0.38671875, | |
| "learning_rate": 7.246762510237404e-06, | |
| "loss": 0.0123, | |
| "step": 163 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "grad_norm": 0.41015625, | |
| "learning_rate": 6.965730342913011e-06, | |
| "loss": 0.0126, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "grad_norm": 0.30078125, | |
| "learning_rate": 6.687313678244243e-06, | |
| "loss": 0.0117, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "grad_norm": 0.41796875, | |
| "learning_rate": 6.411752507928643e-06, | |
| "loss": 0.0106, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "grad_norm": 0.31640625, | |
| "learning_rate": 6.139284362263185e-06, | |
| "loss": 0.0105, | |
| "step": 167 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.326171875, | |
| "learning_rate": 5.8701441053961185e-06, | |
| "loss": 0.01, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "grad_norm": 0.314453125, | |
| "learning_rate": 5.604563732876989e-06, | |
| "loss": 0.0113, | |
| "step": 169 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "grad_norm": 0.361328125, | |
| "learning_rate": 5.342772171679364e-06, | |
| "loss": 0.0111, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "grad_norm": 0.271484375, | |
| "learning_rate": 5.084995082868658e-06, | |
| "loss": 0.0108, | |
| "step": 171 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "grad_norm": 0.41015625, | |
| "learning_rate": 4.831454667085059e-06, | |
| "loss": 0.0114, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "grad_norm": 0.376953125, | |
| "learning_rate": 4.58236947300939e-06, | |
| "loss": 0.012, | |
| "step": 173 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "grad_norm": 0.306640625, | |
| "learning_rate": 4.33795420897683e-06, | |
| "loss": 0.0102, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 0.349609375, | |
| "learning_rate": 4.098419557901036e-06, | |
| "loss": 0.0106, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "eval_loss": 0.02010701596736908, | |
| "eval_runtime": 19.6652, | |
| "eval_samples_per_second": 199.795, | |
| "eval_steps_per_second": 199.795, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "grad_norm": 0.353515625, | |
| "learning_rate": 3.8639719956680624e-06, | |
| "loss": 0.0119, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "grad_norm": 0.400390625, | |
| "learning_rate": 3.6348136131567537e-06, | |
| "loss": 0.0112, | |
| "step": 177 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "grad_norm": 0.3203125, | |
| "learning_rate": 3.4111419420388904e-06, | |
| "loss": 0.0114, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "grad_norm": 0.33203125, | |
| "learning_rate": 3.1931497845093753e-06, | |
| "loss": 0.0109, | |
| "step": 179 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "grad_norm": 0.291015625, | |
| "learning_rate": 2.981025047093118e-06, | |
| "loss": 0.0105, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "grad_norm": 0.296875, | |
| "learning_rate": 2.77495057867198e-06, | |
| "loss": 0.0109, | |
| "step": 181 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "grad_norm": 0.302734375, | |
| "learning_rate": 2.57510401287128e-06, | |
| "loss": 0.0112, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "grad_norm": 0.29296875, | |
| "learning_rate": 2.381657614941858e-06, | |
| "loss": 0.0101, | |
| "step": 183 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "grad_norm": 0.359375, | |
| "learning_rate": 2.1947781332695406e-06, | |
| "loss": 0.0109, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "grad_norm": 0.296875, | |
| "learning_rate": 2.0146266556401405e-06, | |
| "loss": 0.0103, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "grad_norm": 0.27734375, | |
| "learning_rate": 1.8413584703837618e-06, | |
| "loss": 0.0102, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "grad_norm": 0.333984375, | |
| "learning_rate": 1.6751229325182194e-06, | |
| "loss": 0.0101, | |
| "step": 187 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "grad_norm": 0.296875, | |
| "learning_rate": 1.516063335006851e-06, | |
| "loss": 0.0107, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "grad_norm": 0.328125, | |
| "learning_rate": 1.3643167852417894e-06, | |
| "loss": 0.0101, | |
| "step": 189 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "grad_norm": 0.3046875, | |
| "learning_rate": 1.2200140868590759e-06, | |
| "loss": 0.0111, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "grad_norm": 0.2470703125, | |
| "learning_rate": 1.0832796269875757e-06, | |
| "loss": 0.0098, | |
| "step": 191 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "grad_norm": 0.26953125, | |
| "learning_rate": 9.542312690288035e-07, | |
| "loss": 0.0113, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "grad_norm": 0.3671875, | |
| "learning_rate": 8.329802510601559e-07, | |
| "loss": 0.0128, | |
| "step": 193 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "grad_norm": 0.421875, | |
| "learning_rate": 7.196310899490577e-07, | |
| "loss": 0.0114, | |
| "step": 194 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "grad_norm": 0.296875, | |
| "learning_rate": 6.142814912607409e-07, | |
| "loss": 0.0111, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 0.2490234375, | |
| "learning_rate": 5.17022265037247e-07, | |
| "loss": 0.0102, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "grad_norm": 0.28125, | |
| "learning_rate": 4.2793724752031807e-07, | |
| "loss": 0.0089, | |
| "step": 197 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "grad_norm": 0.28515625, | |
| "learning_rate": 3.471032288855869e-07, | |
| "loss": 0.0108, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "grad_norm": 0.29296875, | |
| "learning_rate": 2.745898870504116e-07, | |
| "loss": 0.0092, | |
| "step": 199 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "grad_norm": 0.3203125, | |
| "learning_rate": 2.104597276123721e-07, | |
| "loss": 0.0096, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.267578125, | |
| "learning_rate": 1.5476802997022812e-07, | |
| "loss": 0.0112, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "grad_norm": 0.29296875, | |
| "learning_rate": 1.075627996737627e-07, | |
| "loss": 0.0107, | |
| "step": 202 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "grad_norm": 0.30859375, | |
| "learning_rate": 6.888472704359661e-08, | |
| "loss": 0.011, | |
| "step": 203 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "grad_norm": 0.283203125, | |
| "learning_rate": 3.8767152096641504e-08, | |
| "loss": 0.0099, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "grad_norm": 0.251953125, | |
| "learning_rate": 1.7236035807416397e-08, | |
| "loss": 0.0096, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "grad_norm": 0.291015625, | |
| "learning_rate": 4.309937730015978e-09, | |
| "loss": 0.0106, | |
| "step": 206 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "grad_norm": 0.259765625, | |
| "learning_rate": 0.0, | |
| "loss": 0.0082, | |
| "step": 207 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 207, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 69, | |
| "total_flos": 6.108686871838065e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |