| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 48, | |
| "global_step": 190, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 51.5, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 5.2368, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "eval_loss": 4.782614707946777, | |
| "eval_runtime": 50.8138, | |
| "eval_samples_per_second": 20.742, | |
| "eval_steps_per_second": 20.742, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 48.25, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 4.8965, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 49.25, | |
| "learning_rate": 6e-06, | |
| "loss": 5.0969, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 51.5, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 5.0122, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 48.75, | |
| "learning_rate": 1e-05, | |
| "loss": 5.2505, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 45.25, | |
| "learning_rate": 1.2e-05, | |
| "loss": 4.8966, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 37.5, | |
| "learning_rate": 1.4000000000000001e-05, | |
| "loss": 4.4979, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 30.125, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 4.2038, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 29.625, | |
| "learning_rate": 1.8e-05, | |
| "loss": 4.3498, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 27.5, | |
| "learning_rate": 2e-05, | |
| "loss": 4.0182, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 29.875, | |
| "learning_rate": 2.2000000000000003e-05, | |
| "loss": 3.6018, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 27.375, | |
| "learning_rate": 2.4e-05, | |
| "loss": 3.4676, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 33.5, | |
| "learning_rate": 2.6000000000000002e-05, | |
| "loss": 3.1663, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 20.0, | |
| "learning_rate": 2.8000000000000003e-05, | |
| "loss": 3.0102, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 15.5, | |
| "learning_rate": 3e-05, | |
| "loss": 2.9281, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 17.75, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 2.7986, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 11.0, | |
| "learning_rate": 3.4000000000000007e-05, | |
| "loss": 2.7211, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 15.5, | |
| "learning_rate": 3.6e-05, | |
| "loss": 2.5806, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 17.375, | |
| "learning_rate": 3.8e-05, | |
| "loss": 2.4546, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 10.0625, | |
| "learning_rate": 4e-05, | |
| "loss": 2.6552, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 11.0625, | |
| "learning_rate": 4.2e-05, | |
| "loss": 2.4723, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 10.375, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 2.5615, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 12.5, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 2.4721, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 8.75, | |
| "learning_rate": 4.8e-05, | |
| "loss": 2.3344, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 14.25, | |
| "learning_rate": 5e-05, | |
| "loss": 2.4028, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 9.5625, | |
| "learning_rate": 5.2000000000000004e-05, | |
| "loss": 2.3864, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 11.75, | |
| "learning_rate": 5.4000000000000005e-05, | |
| "loss": 2.3027, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 11.5625, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 2.4142, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 8.5625, | |
| "learning_rate": 5.8e-05, | |
| "loss": 2.3577, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 10.8125, | |
| "learning_rate": 6e-05, | |
| "loss": 2.2604, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 12.0625, | |
| "learning_rate": 6.2e-05, | |
| "loss": 2.1379, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 12.875, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 2.2754, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 14.1875, | |
| "learning_rate": 6.6e-05, | |
| "loss": 2.1848, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 12.5625, | |
| "learning_rate": 6.800000000000001e-05, | |
| "loss": 2.2751, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 10.125, | |
| "learning_rate": 7e-05, | |
| "loss": 2.2342, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 10.3125, | |
| "learning_rate": 7.2e-05, | |
| "loss": 2.1432, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 15.5, | |
| "learning_rate": 7.4e-05, | |
| "loss": 2.4477, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 13.0625, | |
| "learning_rate": 7.6e-05, | |
| "loss": 2.4045, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 9.125, | |
| "learning_rate": 7.800000000000001e-05, | |
| "loss": 2.4156, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 10.5, | |
| "learning_rate": 8e-05, | |
| "loss": 2.2679, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 11.75, | |
| "learning_rate": 8.2e-05, | |
| "loss": 2.3045, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 13.5, | |
| "learning_rate": 8.4e-05, | |
| "loss": 2.0572, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 13.75, | |
| "learning_rate": 8.6e-05, | |
| "loss": 2.1523, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 11.875, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 2.1531, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 11.6875, | |
| "learning_rate": 9e-05, | |
| "loss": 1.9768, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 8.125, | |
| "learning_rate": 9.200000000000001e-05, | |
| "loss": 2.4555, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 6.625, | |
| "learning_rate": 9.4e-05, | |
| "loss": 2.3451, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 16.5, | |
| "learning_rate": 9.6e-05, | |
| "loss": 2.2757, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 2.3981776237487793, | |
| "eval_runtime": 50.6747, | |
| "eval_samples_per_second": 20.799, | |
| "eval_steps_per_second": 20.799, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 16.25, | |
| "learning_rate": 9.8e-05, | |
| "loss": 2.4249, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 5.84375, | |
| "learning_rate": 0.0001, | |
| "loss": 2.306, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 7.75, | |
| "learning_rate": 0.00010200000000000001, | |
| "loss": 2.4711, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 9.1875, | |
| "learning_rate": 0.00010400000000000001, | |
| "loss": 2.3926, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 13.125, | |
| "learning_rate": 0.00010600000000000002, | |
| "loss": 2.1229, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 9.8125, | |
| "learning_rate": 0.00010800000000000001, | |
| "loss": 2.0945, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 12.375, | |
| "learning_rate": 0.00011000000000000002, | |
| "loss": 2.248, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 8.125, | |
| "learning_rate": 0.00011200000000000001, | |
| "loss": 2.349, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 15.5625, | |
| "learning_rate": 0.00011399999999999999, | |
| "loss": 2.2546, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 13.5, | |
| "learning_rate": 0.000116, | |
| "loss": 2.4298, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 8.3125, | |
| "learning_rate": 0.000118, | |
| "loss": 2.1131, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 8.625, | |
| "learning_rate": 0.00012, | |
| "loss": 2.2696, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 6.21875, | |
| "learning_rate": 0.000122, | |
| "loss": 2.244, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 5.875, | |
| "learning_rate": 0.000124, | |
| "loss": 2.1473, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 7.28125, | |
| "learning_rate": 0.000126, | |
| "loss": 2.2231, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 5.1875, | |
| "learning_rate": 0.00012800000000000002, | |
| "loss": 2.0796, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 7.59375, | |
| "learning_rate": 0.00013000000000000002, | |
| "loss": 2.3191, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 6.53125, | |
| "learning_rate": 0.000132, | |
| "loss": 2.214, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 8.375, | |
| "learning_rate": 0.000134, | |
| "loss": 2.3527, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 7.09375, | |
| "learning_rate": 0.00013600000000000003, | |
| "loss": 2.3337, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 7.28125, | |
| "learning_rate": 0.000138, | |
| "loss": 2.2933, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 19.75, | |
| "learning_rate": 0.00014, | |
| "loss": 2.3885, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 16.625, | |
| "learning_rate": 0.000142, | |
| "loss": 2.348, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 6.0, | |
| "learning_rate": 0.000144, | |
| "loss": 2.2594, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 11.875, | |
| "learning_rate": 0.000146, | |
| "loss": 2.5611, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "grad_norm": 5.875, | |
| "learning_rate": 0.000148, | |
| "loss": 2.0386, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "grad_norm": 16.0, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 2.656, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 8.1875, | |
| "learning_rate": 0.000152, | |
| "loss": 2.3885, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 11.375, | |
| "learning_rate": 0.000154, | |
| "loss": 2.701, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 8.75, | |
| "learning_rate": 0.00015600000000000002, | |
| "loss": 2.5388, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 7.84375, | |
| "learning_rate": 0.00015800000000000002, | |
| "loss": 2.5263, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 6.3125, | |
| "learning_rate": 0.00016, | |
| "loss": 2.7169, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 19.125, | |
| "learning_rate": 0.000162, | |
| "loss": 2.504, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 8.25, | |
| "learning_rate": 0.000164, | |
| "loss": 2.4614, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 5.5625, | |
| "learning_rate": 0.000166, | |
| "loss": 2.7586, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 8.3125, | |
| "learning_rate": 0.000168, | |
| "loss": 2.6373, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 8.0625, | |
| "learning_rate": 0.00017, | |
| "loss": 2.3237, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 8.4375, | |
| "learning_rate": 0.000172, | |
| "loss": 2.1896, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 11.5625, | |
| "learning_rate": 0.000174, | |
| "loss": 2.5089, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 7.15625, | |
| "learning_rate": 0.00017600000000000002, | |
| "loss": 2.5806, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 8.875, | |
| "learning_rate": 0.00017800000000000002, | |
| "loss": 2.5497, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 8.6875, | |
| "learning_rate": 0.00018, | |
| "loss": 2.3526, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 5.96875, | |
| "learning_rate": 0.000182, | |
| "loss": 2.283, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 6.375, | |
| "learning_rate": 0.00018400000000000003, | |
| "loss": 2.5388, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 5.5625, | |
| "learning_rate": 0.00018600000000000002, | |
| "loss": 2.4216, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 5.5625, | |
| "learning_rate": 0.000188, | |
| "loss": 2.4199, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 6.5, | |
| "learning_rate": 0.00019, | |
| "loss": 2.6459, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 5.46875, | |
| "learning_rate": 0.000192, | |
| "loss": 2.5869, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_loss": 2.5119166374206543, | |
| "eval_runtime": 50.7277, | |
| "eval_samples_per_second": 20.778, | |
| "eval_steps_per_second": 20.778, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 5.3125, | |
| "learning_rate": 0.000194, | |
| "loss": 2.6162, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 6.5625, | |
| "learning_rate": 0.000196, | |
| "loss": 2.605, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 4.9375, | |
| "learning_rate": 0.00019800000000000002, | |
| "loss": 2.4971, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 8.8125, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4572, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 5.96875, | |
| "learning_rate": 0.0001999390827019096, | |
| "loss": 2.4781, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 9.3125, | |
| "learning_rate": 0.00019975640502598244, | |
| "loss": 2.5578, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 5.625, | |
| "learning_rate": 0.00019945218953682734, | |
| "loss": 2.4608, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 6.65625, | |
| "learning_rate": 0.00019902680687415705, | |
| "loss": 2.874, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 10.875, | |
| "learning_rate": 0.00019848077530122083, | |
| "loss": 2.8733, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 1568.0, | |
| "learning_rate": 0.00019781476007338058, | |
| "loss": 6.3646, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 648.0, | |
| "learning_rate": 0.00019702957262759965, | |
| "loss": 3.75, | |
| "step": 107 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 11.0625, | |
| "learning_rate": 0.0001961261695938319, | |
| "loss": 2.8153, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 6.375, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 2.7939, | |
| "step": 109 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 8.25, | |
| "learning_rate": 0.00019396926207859084, | |
| "loss": 2.8736, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 7.21875, | |
| "learning_rate": 0.00019271838545667876, | |
| "loss": 2.4797, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 5.6875, | |
| "learning_rate": 0.0001913545457642601, | |
| "loss": 2.7665, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 6.625, | |
| "learning_rate": 0.0001898794046299167, | |
| "loss": 2.6241, | |
| "step": 113 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 4.65625, | |
| "learning_rate": 0.00018829475928589271, | |
| "loss": 2.5569, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 6.6875, | |
| "learning_rate": 0.00018660254037844388, | |
| "loss": 2.6452, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 4.6875, | |
| "learning_rate": 0.0001848048096156426, | |
| "loss": 2.6205, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 6.53125, | |
| "learning_rate": 0.00018290375725550417, | |
| "loss": 2.5831, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 5.90625, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 2.6601, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 6.6875, | |
| "learning_rate": 0.00017880107536067218, | |
| "loss": 2.5187, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 5.53125, | |
| "learning_rate": 0.0001766044443118978, | |
| "loss": 2.5023, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 7.84375, | |
| "learning_rate": 0.00017431448254773944, | |
| "loss": 2.6064, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 7.65625, | |
| "learning_rate": 0.0001719339800338651, | |
| "loss": 2.5065, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 6.15625, | |
| "learning_rate": 0.00016946583704589973, | |
| "loss": 2.6022, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 5.15625, | |
| "learning_rate": 0.00016691306063588583, | |
| "loss": 2.5981, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 5.28125, | |
| "learning_rate": 0.00016427876096865394, | |
| "loss": 2.6804, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 5.71875, | |
| "learning_rate": 0.0001615661475325658, | |
| "loss": 2.7233, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 3.84375, | |
| "learning_rate": 0.00015877852522924732, | |
| "loss": 2.4712, | |
| "step": 127 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 5.9375, | |
| "learning_rate": 0.0001559192903470747, | |
| "loss": 2.4518, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 4.53125, | |
| "learning_rate": 0.0001529919264233205, | |
| "loss": 2.5456, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 5.4375, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 2.4622, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "grad_norm": 4.84375, | |
| "learning_rate": 0.00014694715627858908, | |
| "loss": 2.6133, | |
| "step": 131 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "grad_norm": 5.65625, | |
| "learning_rate": 0.00014383711467890774, | |
| "loss": 2.4844, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 5.09375, | |
| "learning_rate": 0.00014067366430758004, | |
| "loss": 2.676, | |
| "step": 133 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 6.65625, | |
| "learning_rate": 0.00013746065934159123, | |
| "loss": 2.6896, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 5.5625, | |
| "learning_rate": 0.00013420201433256689, | |
| "loss": 2.5363, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 4.34375, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 2.2558, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 6.53125, | |
| "learning_rate": 0.0001275637355816999, | |
| "loss": 2.5952, | |
| "step": 137 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 4.28125, | |
| "learning_rate": 0.00012419218955996676, | |
| "loss": 2.3442, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 4.28125, | |
| "learning_rate": 0.00012079116908177593, | |
| "loss": 2.4271, | |
| "step": 139 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 4.28125, | |
| "learning_rate": 0.00011736481776669306, | |
| "loss": 2.4587, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 3.671875, | |
| "learning_rate": 0.00011391731009600654, | |
| "loss": 2.5501, | |
| "step": 141 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 5.375, | |
| "learning_rate": 0.00011045284632676536, | |
| "loss": 2.4936, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 5.5625, | |
| "learning_rate": 0.00010697564737441252, | |
| "loss": 2.1856, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 4.03125, | |
| "learning_rate": 0.00010348994967025012, | |
| "loss": 2.664, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "eval_loss": 2.378908395767212, | |
| "eval_runtime": 50.7719, | |
| "eval_samples_per_second": 20.76, | |
| "eval_steps_per_second": 20.76, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 5.15625, | |
| "learning_rate": 0.0001, | |
| "loss": 2.4778, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 4.625, | |
| "learning_rate": 9.651005032974994e-05, | |
| "loss": 2.388, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 4.1875, | |
| "learning_rate": 9.302435262558747e-05, | |
| "loss": 2.4377, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 6.40625, | |
| "learning_rate": 8.954715367323468e-05, | |
| "loss": 2.5183, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 5.09375, | |
| "learning_rate": 8.608268990399349e-05, | |
| "loss": 2.5759, | |
| "step": 149 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 4.4375, | |
| "learning_rate": 8.263518223330697e-05, | |
| "loss": 2.6291, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 5.0625, | |
| "learning_rate": 7.920883091822408e-05, | |
| "loss": 2.3462, | |
| "step": 151 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 3.875, | |
| "learning_rate": 7.580781044003324e-05, | |
| "loss": 2.1391, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 3.625, | |
| "learning_rate": 7.243626441830009e-05, | |
| "loss": 2.3882, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 4.21875, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 2.3264, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 3.90625, | |
| "learning_rate": 6.579798566743314e-05, | |
| "loss": 2.2026, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 5.75, | |
| "learning_rate": 6.25393406584088e-05, | |
| "loss": 2.5855, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "grad_norm": 4.6875, | |
| "learning_rate": 5.9326335692419995e-05, | |
| "loss": 2.2483, | |
| "step": 157 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "grad_norm": 3.65625, | |
| "learning_rate": 5.616288532109225e-05, | |
| "loss": 2.1449, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 3.578125, | |
| "learning_rate": 5.305284372141095e-05, | |
| "loss": 2.2383, | |
| "step": 159 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 3.65625, | |
| "learning_rate": 5.000000000000002e-05, | |
| "loss": 2.2851, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 4.125, | |
| "learning_rate": 4.700807357667952e-05, | |
| "loss": 2.4617, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 4.4375, | |
| "learning_rate": 4.4080709652925336e-05, | |
| "loss": 2.1831, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 4.3125, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 2.2079, | |
| "step": 163 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 3.28125, | |
| "learning_rate": 3.843385246743417e-05, | |
| "loss": 2.309, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 5.15625, | |
| "learning_rate": 3.5721239031346066e-05, | |
| "loss": 2.4679, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 5.09375, | |
| "learning_rate": 3.308693936411421e-05, | |
| "loss": 2.3249, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 4.625, | |
| "learning_rate": 3.053416295410026e-05, | |
| "loss": 2.2908, | |
| "step": 167 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 3.90625, | |
| "learning_rate": 2.8066019966134904e-05, | |
| "loss": 2.3121, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 3.8125, | |
| "learning_rate": 2.5685517452260567e-05, | |
| "loss": 2.2577, | |
| "step": 169 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 3.453125, | |
| "learning_rate": 2.339555568810221e-05, | |
| "loss": 2.2002, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 3.84375, | |
| "learning_rate": 2.119892463932781e-05, | |
| "loss": 2.4345, | |
| "step": 171 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 4.21875, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 2.2662, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 3.546875, | |
| "learning_rate": 1.7096242744495837e-05, | |
| "loss": 2.4411, | |
| "step": 173 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 3.640625, | |
| "learning_rate": 1.5195190384357404e-05, | |
| "loss": 2.4609, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 4.375, | |
| "learning_rate": 1.339745962155613e-05, | |
| "loss": 2.3615, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 3.4375, | |
| "learning_rate": 1.1705240714107302e-05, | |
| "loss": 2.2071, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 3.578125, | |
| "learning_rate": 1.0120595370083318e-05, | |
| "loss": 2.3504, | |
| "step": 177 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 3.578125, | |
| "learning_rate": 8.645454235739903e-06, | |
| "loss": 2.2771, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 3.5625, | |
| "learning_rate": 7.281614543321269e-06, | |
| "loss": 2.3663, | |
| "step": 179 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 3.234375, | |
| "learning_rate": 6.030737921409169e-06, | |
| "loss": 2.4842, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 3.5625, | |
| "learning_rate": 4.8943483704846475e-06, | |
| "loss": 2.3921, | |
| "step": 181 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 3.484375, | |
| "learning_rate": 3.873830406168111e-06, | |
| "loss": 2.3467, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 3.34375, | |
| "learning_rate": 2.970427372400353e-06, | |
| "loss": 2.3193, | |
| "step": 183 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 3.71875, | |
| "learning_rate": 2.1852399266194314e-06, | |
| "loss": 2.3215, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 3.5625, | |
| "learning_rate": 1.5192246987791981e-06, | |
| "loss": 2.1687, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "grad_norm": 3.359375, | |
| "learning_rate": 9.731931258429638e-07, | |
| "loss": 2.2471, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "grad_norm": 3.203125, | |
| "learning_rate": 5.478104631726711e-07, | |
| "loss": 2.356, | |
| "step": 187 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 3.25, | |
| "learning_rate": 2.4359497401758024e-07, | |
| "loss": 2.2989, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 3.15625, | |
| "learning_rate": 6.09172980904238e-08, | |
| "loss": 2.0376, | |
| "step": 189 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 3.625, | |
| "learning_rate": 0.0, | |
| "loss": 2.5224, | |
| "step": 190 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 190, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "total_flos": 2415296753172480.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |