| { | |
| "best_metric": 1.810941219329834, | |
| "best_model_checkpoint": "lora_lr/google/gemma-1.1-7b-it/unaligned/checkpoint-250", | |
| "epoch": 0.327575175943698, | |
| "eval_steps": 10, | |
| "global_step": 256, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 1.015625, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 4.7488, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 0.94140625, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 4.5149, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 0.9140625, | |
| "learning_rate": 6e-06, | |
| "loss": 4.3894, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 1.0078125, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 4.6734, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 1.0, | |
| "learning_rate": 1e-05, | |
| "loss": 4.6616, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.921875, | |
| "learning_rate": 1.2e-05, | |
| "loss": 4.5625, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.9765625, | |
| "learning_rate": 1.4000000000000001e-05, | |
| "loss": 4.5709, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.99609375, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 4.6627, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 1.0390625, | |
| "learning_rate": 1.8e-05, | |
| "loss": 4.5995, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 1.0390625, | |
| "learning_rate": 2e-05, | |
| "loss": 4.7668, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "eval_loss": 4.575911521911621, | |
| "eval_runtime": 126.7034, | |
| "eval_samples_per_second": 39.462, | |
| "eval_steps_per_second": 1.239, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 1.03125, | |
| "learning_rate": 2.2000000000000003e-05, | |
| "loss": 4.5432, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.9296875, | |
| "learning_rate": 2.4e-05, | |
| "loss": 4.2969, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 1.1015625, | |
| "learning_rate": 2.6000000000000002e-05, | |
| "loss": 4.8134, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 1.109375, | |
| "learning_rate": 2.8000000000000003e-05, | |
| "loss": 4.6818, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 1.1171875, | |
| "learning_rate": 3e-05, | |
| "loss": 4.5194, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 1.1015625, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 4.3755, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 1.265625, | |
| "learning_rate": 3.4000000000000007e-05, | |
| "loss": 4.5325, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 1.2265625, | |
| "learning_rate": 3.6e-05, | |
| "loss": 4.3967, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 1.359375, | |
| "learning_rate": 3.8e-05, | |
| "loss": 4.3274, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 1.375, | |
| "learning_rate": 4e-05, | |
| "loss": 4.3029, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "eval_loss": 4.288708686828613, | |
| "eval_runtime": 126.1277, | |
| "eval_samples_per_second": 39.642, | |
| "eval_steps_per_second": 1.245, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 1.390625, | |
| "learning_rate": 4.2e-05, | |
| "loss": 4.2266, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 1.5625, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 4.3114, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 1.609375, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 4.2829, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 1.6796875, | |
| "learning_rate": 4.8e-05, | |
| "loss": 4.1797, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 1.75, | |
| "learning_rate": 5e-05, | |
| "loss": 4.0528, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 1.8359375, | |
| "learning_rate": 5.2000000000000004e-05, | |
| "loss": 3.997, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 1.828125, | |
| "learning_rate": 5.4000000000000005e-05, | |
| "loss": 3.9195, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 1.8203125, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 3.817, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 1.9296875, | |
| "learning_rate": 5.8e-05, | |
| "loss": 3.8411, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 1.921875, | |
| "learning_rate": 6e-05, | |
| "loss": 3.5259, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "eval_loss": 3.5061299800872803, | |
| "eval_runtime": 126.2241, | |
| "eval_samples_per_second": 39.612, | |
| "eval_steps_per_second": 1.244, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 2.0, | |
| "learning_rate": 6.2e-05, | |
| "loss": 3.6587, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 1.765625, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 3.3672, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 1.75, | |
| "learning_rate": 6.6e-05, | |
| "loss": 3.3194, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 1.703125, | |
| "learning_rate": 6.800000000000001e-05, | |
| "loss": 3.2215, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 1.6640625, | |
| "learning_rate": 7e-05, | |
| "loss": 3.1773, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 1.546875, | |
| "learning_rate": 7.2e-05, | |
| "loss": 2.9527, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 1.578125, | |
| "learning_rate": 7.4e-05, | |
| "loss": 3.1381, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 1.453125, | |
| "learning_rate": 7.6e-05, | |
| "loss": 2.947, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 1.2265625, | |
| "learning_rate": 7.800000000000001e-05, | |
| "loss": 2.743, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 1.1171875, | |
| "learning_rate": 8e-05, | |
| "loss": 2.6947, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "eval_loss": 2.6360573768615723, | |
| "eval_runtime": 125.9592, | |
| "eval_samples_per_second": 39.695, | |
| "eval_steps_per_second": 1.246, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.9453125, | |
| "learning_rate": 8.2e-05, | |
| "loss": 2.6962, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 1.0390625, | |
| "learning_rate": 8.4e-05, | |
| "loss": 2.6634, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 1.3203125, | |
| "learning_rate": 8.6e-05, | |
| "loss": 2.5018, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 1.390625, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 2.4608, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 1.328125, | |
| "learning_rate": 9e-05, | |
| "loss": 2.425, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.5859375, | |
| "learning_rate": 9.200000000000001e-05, | |
| "loss": 2.3978, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.5625, | |
| "learning_rate": 9.4e-05, | |
| "loss": 2.3885, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.56640625, | |
| "learning_rate": 9.6e-05, | |
| "loss": 2.373, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.64453125, | |
| "learning_rate": 9.8e-05, | |
| "loss": 2.2775, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.609375, | |
| "learning_rate": 0.0001, | |
| "loss": 2.3577, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "eval_loss": 2.296643018722534, | |
| "eval_runtime": 125.5906, | |
| "eval_samples_per_second": 39.812, | |
| "eval_steps_per_second": 1.25, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.5703125, | |
| "learning_rate": 0.00010200000000000001, | |
| "loss": 2.447, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.59765625, | |
| "learning_rate": 0.00010400000000000001, | |
| "loss": 2.2502, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.5546875, | |
| "learning_rate": 0.00010600000000000002, | |
| "loss": 2.2093, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.8203125, | |
| "learning_rate": 0.00010800000000000001, | |
| "loss": 2.2644, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 1.1171875, | |
| "learning_rate": 0.00011000000000000002, | |
| "loss": 2.1485, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.92578125, | |
| "learning_rate": 0.00011200000000000001, | |
| "loss": 2.1965, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.70703125, | |
| "learning_rate": 0.00011399999999999999, | |
| "loss": 2.1916, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.53125, | |
| "learning_rate": 0.000116, | |
| "loss": 2.2071, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.392578125, | |
| "learning_rate": 0.000118, | |
| "loss": 2.1507, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.447265625, | |
| "learning_rate": 0.00012, | |
| "loss": 2.1582, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "eval_loss": 2.1068594455718994, | |
| "eval_runtime": 124.951, | |
| "eval_samples_per_second": 40.016, | |
| "eval_steps_per_second": 1.256, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.3984375, | |
| "learning_rate": 0.000122, | |
| "loss": 2.1486, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.38671875, | |
| "learning_rate": 0.000124, | |
| "loss": 2.1137, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.65625, | |
| "learning_rate": 0.000126, | |
| "loss": 2.1276, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.43359375, | |
| "learning_rate": 0.00012800000000000002, | |
| "loss": 2.0382, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.5546875, | |
| "learning_rate": 0.00013000000000000002, | |
| "loss": 2.118, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.94921875, | |
| "learning_rate": 0.000132, | |
| "loss": 2.1625, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.443359375, | |
| "learning_rate": 0.000134, | |
| "loss": 2.0544, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.36328125, | |
| "learning_rate": 0.00013600000000000003, | |
| "loss": 2.0529, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.353515625, | |
| "learning_rate": 0.000138, | |
| "loss": 2.0054, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.35546875, | |
| "learning_rate": 0.00014, | |
| "loss": 2.0338, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "eval_loss": 2.0194942951202393, | |
| "eval_runtime": 125.4346, | |
| "eval_samples_per_second": 39.861, | |
| "eval_steps_per_second": 1.252, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.384765625, | |
| "learning_rate": 0.000142, | |
| "loss": 2.0377, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.3515625, | |
| "learning_rate": 0.000144, | |
| "loss": 2.0574, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.375, | |
| "learning_rate": 0.000146, | |
| "loss": 1.9926, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.322265625, | |
| "learning_rate": 0.000148, | |
| "loss": 2.1027, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.33984375, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 2.0283, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.28125, | |
| "learning_rate": 0.000152, | |
| "loss": 1.9844, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.291015625, | |
| "learning_rate": 0.000154, | |
| "loss": 2.0037, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.23046875, | |
| "learning_rate": 0.00015600000000000002, | |
| "loss": 1.9911, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.353515625, | |
| "learning_rate": 0.00015800000000000002, | |
| "loss": 1.933, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.69140625, | |
| "learning_rate": 0.00016, | |
| "loss": 1.9856, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_loss": 1.9626843929290771, | |
| "eval_runtime": 125.5025, | |
| "eval_samples_per_second": 39.84, | |
| "eval_steps_per_second": 1.251, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.26171875, | |
| "learning_rate": 0.000162, | |
| "loss": 1.9761, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.2734375, | |
| "learning_rate": 0.000164, | |
| "loss": 1.9205, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.279296875, | |
| "learning_rate": 0.000166, | |
| "loss": 1.9885, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.291015625, | |
| "learning_rate": 0.000168, | |
| "loss": 1.9419, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.298828125, | |
| "learning_rate": 0.00017, | |
| "loss": 1.9895, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.27734375, | |
| "learning_rate": 0.000172, | |
| "loss": 1.8674, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.25390625, | |
| "learning_rate": 0.000174, | |
| "loss": 1.9472, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.361328125, | |
| "learning_rate": 0.00017600000000000002, | |
| "loss": 1.9717, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.244140625, | |
| "learning_rate": 0.00017800000000000002, | |
| "loss": 1.9376, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.234375, | |
| "learning_rate": 0.00018, | |
| "loss": 1.9051, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "eval_loss": 1.9230234622955322, | |
| "eval_runtime": 125.5293, | |
| "eval_samples_per_second": 39.831, | |
| "eval_steps_per_second": 1.251, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.26171875, | |
| "learning_rate": 0.000182, | |
| "loss": 1.9395, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.310546875, | |
| "learning_rate": 0.00018400000000000003, | |
| "loss": 1.8496, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.3203125, | |
| "learning_rate": 0.00018600000000000002, | |
| "loss": 2.0062, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.267578125, | |
| "learning_rate": 0.000188, | |
| "loss": 1.9138, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.224609375, | |
| "learning_rate": 0.00019, | |
| "loss": 1.9258, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.21484375, | |
| "learning_rate": 0.000192, | |
| "loss": 1.9657, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.259765625, | |
| "learning_rate": 0.000194, | |
| "loss": 1.9123, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 0.240234375, | |
| "learning_rate": 0.000196, | |
| "loss": 1.8806, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 0.21484375, | |
| "learning_rate": 0.00019800000000000002, | |
| "loss": 1.8394, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 0.244140625, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8164, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "eval_loss": 1.8982120752334595, | |
| "eval_runtime": 125.467, | |
| "eval_samples_per_second": 39.851, | |
| "eval_steps_per_second": 1.251, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 0.265625, | |
| "learning_rate": 0.00019871794871794874, | |
| "loss": 1.8199, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 0.2158203125, | |
| "learning_rate": 0.00019743589743589744, | |
| "loss": 1.9268, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 0.2578125, | |
| "learning_rate": 0.00019615384615384615, | |
| "loss": 1.9563, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 0.20703125, | |
| "learning_rate": 0.00019487179487179487, | |
| "loss": 1.9037, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 0.2265625, | |
| "learning_rate": 0.0001935897435897436, | |
| "loss": 1.8942, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.2158203125, | |
| "learning_rate": 0.00019230769230769233, | |
| "loss": 1.9245, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.2431640625, | |
| "learning_rate": 0.00019102564102564104, | |
| "loss": 1.8898, | |
| "step": 107 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.310546875, | |
| "learning_rate": 0.00018974358974358974, | |
| "loss": 1.8482, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.2197265625, | |
| "learning_rate": 0.00018846153846153847, | |
| "loss": 1.8486, | |
| "step": 109 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.228515625, | |
| "learning_rate": 0.0001871794871794872, | |
| "loss": 1.9404, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "eval_loss": 1.877401351928711, | |
| "eval_runtime": 125.8694, | |
| "eval_samples_per_second": 39.724, | |
| "eval_steps_per_second": 1.247, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.2734375, | |
| "learning_rate": 0.0001858974358974359, | |
| "loss": 1.8611, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.2216796875, | |
| "learning_rate": 0.00018461538461538463, | |
| "loss": 1.8728, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.27734375, | |
| "learning_rate": 0.00018333333333333334, | |
| "loss": 1.8817, | |
| "step": 113 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.224609375, | |
| "learning_rate": 0.00018205128205128207, | |
| "loss": 1.8841, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.2294921875, | |
| "learning_rate": 0.00018076923076923077, | |
| "loss": 1.8691, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.28515625, | |
| "learning_rate": 0.0001794871794871795, | |
| "loss": 1.9013, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.2265625, | |
| "learning_rate": 0.00017820512820512823, | |
| "loss": 1.9729, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.2080078125, | |
| "learning_rate": 0.00017692307692307693, | |
| "loss": 1.8934, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.2275390625, | |
| "learning_rate": 0.00017564102564102566, | |
| "loss": 1.8559, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.21875, | |
| "learning_rate": 0.00017435897435897436, | |
| "loss": 1.9225, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 1.8639016151428223, | |
| "eval_runtime": 125.6482, | |
| "eval_samples_per_second": 39.794, | |
| "eval_steps_per_second": 1.25, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.2138671875, | |
| "learning_rate": 0.0001730769230769231, | |
| "loss": 1.8796, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.29296875, | |
| "learning_rate": 0.0001717948717948718, | |
| "loss": 1.8587, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.2236328125, | |
| "learning_rate": 0.00017051282051282053, | |
| "loss": 1.8642, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.240234375, | |
| "learning_rate": 0.00016923076923076923, | |
| "loss": 1.8532, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.2216796875, | |
| "learning_rate": 0.00016794871794871796, | |
| "loss": 1.8731, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.310546875, | |
| "learning_rate": 0.0001666666666666667, | |
| "loss": 1.8667, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.22265625, | |
| "learning_rate": 0.0001653846153846154, | |
| "loss": 1.9161, | |
| "step": 127 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.2236328125, | |
| "learning_rate": 0.0001641025641025641, | |
| "loss": 1.831, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.271484375, | |
| "learning_rate": 0.00016282051282051282, | |
| "loss": 1.9271, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.220703125, | |
| "learning_rate": 0.00016153846153846155, | |
| "loss": 1.9196, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "eval_loss": 1.854244589805603, | |
| "eval_runtime": 125.489, | |
| "eval_samples_per_second": 39.844, | |
| "eval_steps_per_second": 1.251, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.294921875, | |
| "learning_rate": 0.00016025641025641028, | |
| "loss": 1.8143, | |
| "step": 131 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.291015625, | |
| "learning_rate": 0.00015897435897435896, | |
| "loss": 1.8742, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.275390625, | |
| "learning_rate": 0.0001576923076923077, | |
| "loss": 1.9119, | |
| "step": 133 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.2353515625, | |
| "learning_rate": 0.00015641025641025642, | |
| "loss": 1.9378, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.2431640625, | |
| "learning_rate": 0.00015512820512820515, | |
| "loss": 1.8725, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.314453125, | |
| "learning_rate": 0.00015384615384615385, | |
| "loss": 1.8882, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.32421875, | |
| "learning_rate": 0.00015256410256410255, | |
| "loss": 1.7994, | |
| "step": 137 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.244140625, | |
| "learning_rate": 0.00015128205128205128, | |
| "loss": 1.7943, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.248046875, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 1.8982, | |
| "step": 139 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.251953125, | |
| "learning_rate": 0.00014871794871794872, | |
| "loss": 1.8085, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "eval_loss": 1.844925045967102, | |
| "eval_runtime": 125.5735, | |
| "eval_samples_per_second": 39.817, | |
| "eval_steps_per_second": 1.25, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.2490234375, | |
| "learning_rate": 0.00014743589743589745, | |
| "loss": 1.8725, | |
| "step": 141 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.21875, | |
| "learning_rate": 0.00014615384615384615, | |
| "loss": 1.8679, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.2421875, | |
| "learning_rate": 0.00014487179487179488, | |
| "loss": 1.8594, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.2431640625, | |
| "learning_rate": 0.0001435897435897436, | |
| "loss": 1.8896, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.42578125, | |
| "learning_rate": 0.0001423076923076923, | |
| "loss": 1.8578, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.2392578125, | |
| "learning_rate": 0.00014102564102564104, | |
| "loss": 1.8631, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.33203125, | |
| "learning_rate": 0.00013974358974358974, | |
| "loss": 1.8408, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.251953125, | |
| "learning_rate": 0.00013846153846153847, | |
| "loss": 1.752, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.326171875, | |
| "learning_rate": 0.00013717948717948718, | |
| "loss": 1.9149, | |
| "step": 149 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.267578125, | |
| "learning_rate": 0.0001358974358974359, | |
| "loss": 1.9018, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "eval_loss": 1.8380218744277954, | |
| "eval_runtime": 125.5888, | |
| "eval_samples_per_second": 39.812, | |
| "eval_steps_per_second": 1.25, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.28125, | |
| "learning_rate": 0.00013461538461538464, | |
| "loss": 1.7853, | |
| "step": 151 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.2470703125, | |
| "learning_rate": 0.00013333333333333334, | |
| "loss": 1.9392, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.2490234375, | |
| "learning_rate": 0.00013205128205128204, | |
| "loss": 1.8372, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.28515625, | |
| "learning_rate": 0.00013076923076923077, | |
| "loss": 1.8017, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.255859375, | |
| "learning_rate": 0.0001294871794871795, | |
| "loss": 1.8477, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.2197265625, | |
| "learning_rate": 0.00012820512820512823, | |
| "loss": 1.773, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.2412109375, | |
| "learning_rate": 0.00012692307692307693, | |
| "loss": 1.8194, | |
| "step": 157 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.251953125, | |
| "learning_rate": 0.00012564102564102564, | |
| "loss": 1.841, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.2578125, | |
| "learning_rate": 0.00012435897435897437, | |
| "loss": 1.8599, | |
| "step": 159 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.2392578125, | |
| "learning_rate": 0.0001230769230769231, | |
| "loss": 1.8224, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_loss": 1.8314405679702759, | |
| "eval_runtime": 125.9116, | |
| "eval_samples_per_second": 39.71, | |
| "eval_steps_per_second": 1.247, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.26953125, | |
| "learning_rate": 0.00012179487179487179, | |
| "loss": 1.8628, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.244140625, | |
| "learning_rate": 0.00012051282051282052, | |
| "loss": 1.8015, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.283203125, | |
| "learning_rate": 0.00011923076923076923, | |
| "loss": 1.8325, | |
| "step": 163 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.2578125, | |
| "learning_rate": 0.00011794871794871796, | |
| "loss": 1.8038, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.23828125, | |
| "learning_rate": 0.00011666666666666668, | |
| "loss": 1.729, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.375, | |
| "learning_rate": 0.00011538461538461538, | |
| "loss": 1.8223, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.30859375, | |
| "learning_rate": 0.0001141025641025641, | |
| "loss": 1.8585, | |
| "step": 167 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.271484375, | |
| "learning_rate": 0.00011282051282051283, | |
| "loss": 1.9158, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.2734375, | |
| "learning_rate": 0.00011153846153846154, | |
| "loss": 1.8058, | |
| "step": 169 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.265625, | |
| "learning_rate": 0.00011025641025641027, | |
| "loss": 1.9334, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_loss": 1.8285787105560303, | |
| "eval_runtime": 125.7453, | |
| "eval_samples_per_second": 39.763, | |
| "eval_steps_per_second": 1.249, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.376953125, | |
| "learning_rate": 0.00010897435897435896, | |
| "loss": 1.8184, | |
| "step": 171 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.24609375, | |
| "learning_rate": 0.0001076923076923077, | |
| "loss": 1.7568, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.28515625, | |
| "learning_rate": 0.00010641025641025641, | |
| "loss": 1.847, | |
| "step": 173 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.267578125, | |
| "learning_rate": 0.00010512820512820514, | |
| "loss": 1.8164, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.2890625, | |
| "learning_rate": 0.00010384615384615386, | |
| "loss": 1.9219, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.25390625, | |
| "learning_rate": 0.00010256410256410256, | |
| "loss": 1.8789, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.275390625, | |
| "learning_rate": 0.00010128205128205129, | |
| "loss": 1.8036, | |
| "step": 177 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.306640625, | |
| "learning_rate": 0.0001, | |
| "loss": 1.7451, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.27734375, | |
| "learning_rate": 9.871794871794872e-05, | |
| "loss": 1.8295, | |
| "step": 179 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.26171875, | |
| "learning_rate": 9.743589743589744e-05, | |
| "loss": 1.8326, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "eval_loss": 1.8210687637329102, | |
| "eval_runtime": 125.5174, | |
| "eval_samples_per_second": 39.835, | |
| "eval_steps_per_second": 1.251, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.267578125, | |
| "learning_rate": 9.615384615384617e-05, | |
| "loss": 1.817, | |
| "step": 181 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.291015625, | |
| "learning_rate": 9.487179487179487e-05, | |
| "loss": 1.7804, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.267578125, | |
| "learning_rate": 9.35897435897436e-05, | |
| "loss": 1.8063, | |
| "step": 183 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.287109375, | |
| "learning_rate": 9.230769230769232e-05, | |
| "loss": 1.7806, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.275390625, | |
| "learning_rate": 9.102564102564103e-05, | |
| "loss": 1.9115, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.265625, | |
| "learning_rate": 8.974358974358975e-05, | |
| "loss": 1.8414, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.2314453125, | |
| "learning_rate": 8.846153846153847e-05, | |
| "loss": 1.8451, | |
| "step": 187 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.2294921875, | |
| "learning_rate": 8.717948717948718e-05, | |
| "loss": 1.7911, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.30078125, | |
| "learning_rate": 8.58974358974359e-05, | |
| "loss": 1.8654, | |
| "step": 189 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.27734375, | |
| "learning_rate": 8.461538461538461e-05, | |
| "loss": 1.8412, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_loss": 1.8182088136672974, | |
| "eval_runtime": 125.5312, | |
| "eval_samples_per_second": 39.831, | |
| "eval_steps_per_second": 1.251, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.267578125, | |
| "learning_rate": 8.333333333333334e-05, | |
| "loss": 1.869, | |
| "step": 191 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.263671875, | |
| "learning_rate": 8.205128205128205e-05, | |
| "loss": 1.8746, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.25390625, | |
| "learning_rate": 8.076923076923078e-05, | |
| "loss": 1.7365, | |
| "step": 193 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.2734375, | |
| "learning_rate": 7.948717948717948e-05, | |
| "loss": 1.891, | |
| "step": 194 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.248046875, | |
| "learning_rate": 7.820512820512821e-05, | |
| "loss": 1.7924, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.240234375, | |
| "learning_rate": 7.692307692307693e-05, | |
| "loss": 1.8078, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.302734375, | |
| "learning_rate": 7.564102564102564e-05, | |
| "loss": 1.8012, | |
| "step": 197 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.275390625, | |
| "learning_rate": 7.435897435897436e-05, | |
| "loss": 1.8252, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.26953125, | |
| "learning_rate": 7.307692307692307e-05, | |
| "loss": 1.8348, | |
| "step": 199 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.2734375, | |
| "learning_rate": 7.17948717948718e-05, | |
| "loss": 1.7759, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "eval_loss": 1.8161256313323975, | |
| "eval_runtime": 125.4822, | |
| "eval_samples_per_second": 39.846, | |
| "eval_steps_per_second": 1.251, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.265625, | |
| "learning_rate": 7.051282051282052e-05, | |
| "loss": 1.8081, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.2412109375, | |
| "learning_rate": 6.923076923076924e-05, | |
| "loss": 1.8619, | |
| "step": 202 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.263671875, | |
| "learning_rate": 6.794871794871795e-05, | |
| "loss": 1.898, | |
| "step": 203 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.255859375, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 1.8495, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.25, | |
| "learning_rate": 6.538461538461539e-05, | |
| "loss": 1.7347, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.287109375, | |
| "learning_rate": 6.410256410256412e-05, | |
| "loss": 1.8185, | |
| "step": 206 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.251953125, | |
| "learning_rate": 6.282051282051282e-05, | |
| "loss": 1.7143, | |
| "step": 207 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.255859375, | |
| "learning_rate": 6.153846153846155e-05, | |
| "loss": 1.8391, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.3125, | |
| "learning_rate": 6.025641025641026e-05, | |
| "loss": 1.8135, | |
| "step": 209 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.265625, | |
| "learning_rate": 5.897435897435898e-05, | |
| "loss": 1.8639, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "eval_loss": 1.8143140077590942, | |
| "eval_runtime": 125.8394, | |
| "eval_samples_per_second": 39.733, | |
| "eval_steps_per_second": 1.248, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.259765625, | |
| "learning_rate": 5.769230769230769e-05, | |
| "loss": 1.8108, | |
| "step": 211 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.251953125, | |
| "learning_rate": 5.6410256410256414e-05, | |
| "loss": 1.8078, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.27734375, | |
| "learning_rate": 5.512820512820514e-05, | |
| "loss": 1.8167, | |
| "step": 213 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.26171875, | |
| "learning_rate": 5.384615384615385e-05, | |
| "loss": 1.8148, | |
| "step": 214 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.26171875, | |
| "learning_rate": 5.256410256410257e-05, | |
| "loss": 1.7895, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.275390625, | |
| "learning_rate": 5.128205128205128e-05, | |
| "loss": 1.7537, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.283203125, | |
| "learning_rate": 5e-05, | |
| "loss": 1.8267, | |
| "step": 217 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.26171875, | |
| "learning_rate": 4.871794871794872e-05, | |
| "loss": 1.9016, | |
| "step": 218 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.26953125, | |
| "learning_rate": 4.7435897435897435e-05, | |
| "loss": 1.8097, | |
| "step": 219 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.283203125, | |
| "learning_rate": 4.615384615384616e-05, | |
| "loss": 1.8519, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "eval_loss": 1.8137133121490479, | |
| "eval_runtime": 125.6108, | |
| "eval_samples_per_second": 39.805, | |
| "eval_steps_per_second": 1.25, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.2734375, | |
| "learning_rate": 4.4871794871794874e-05, | |
| "loss": 1.8262, | |
| "step": 221 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.271484375, | |
| "learning_rate": 4.358974358974359e-05, | |
| "loss": 1.8572, | |
| "step": 222 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.251953125, | |
| "learning_rate": 4.230769230769231e-05, | |
| "loss": 1.8059, | |
| "step": 223 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.255859375, | |
| "learning_rate": 4.1025641025641023e-05, | |
| "loss": 1.813, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.259765625, | |
| "learning_rate": 3.974358974358974e-05, | |
| "loss": 1.8061, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.26171875, | |
| "learning_rate": 3.846153846153846e-05, | |
| "loss": 1.7375, | |
| "step": 226 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.2734375, | |
| "learning_rate": 3.717948717948718e-05, | |
| "loss": 1.8433, | |
| "step": 227 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.27734375, | |
| "learning_rate": 3.58974358974359e-05, | |
| "loss": 1.8459, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.2578125, | |
| "learning_rate": 3.461538461538462e-05, | |
| "loss": 1.7373, | |
| "step": 229 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.248046875, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 1.862, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": 1.8112460374832153, | |
| "eval_runtime": 125.6176, | |
| "eval_samples_per_second": 39.803, | |
| "eval_steps_per_second": 1.25, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.306640625, | |
| "learning_rate": 3.205128205128206e-05, | |
| "loss": 1.8498, | |
| "step": 231 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.2578125, | |
| "learning_rate": 3.0769230769230774e-05, | |
| "loss": 1.7185, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.25390625, | |
| "learning_rate": 2.948717948717949e-05, | |
| "loss": 1.7511, | |
| "step": 233 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.267578125, | |
| "learning_rate": 2.8205128205128207e-05, | |
| "loss": 1.8457, | |
| "step": 234 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.267578125, | |
| "learning_rate": 2.6923076923076923e-05, | |
| "loss": 1.8456, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.26953125, | |
| "learning_rate": 2.564102564102564e-05, | |
| "loss": 1.8656, | |
| "step": 236 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.2578125, | |
| "learning_rate": 2.435897435897436e-05, | |
| "loss": 1.7922, | |
| "step": 237 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.26953125, | |
| "learning_rate": 2.307692307692308e-05, | |
| "loss": 1.8367, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.2470703125, | |
| "learning_rate": 2.1794871794871795e-05, | |
| "loss": 1.8649, | |
| "step": 239 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.349609375, | |
| "learning_rate": 2.0512820512820512e-05, | |
| "loss": 1.8575, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eval_loss": 1.811773419380188, | |
| "eval_runtime": 125.6114, | |
| "eval_samples_per_second": 39.805, | |
| "eval_steps_per_second": 1.25, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.251953125, | |
| "learning_rate": 1.923076923076923e-05, | |
| "loss": 1.8488, | |
| "step": 241 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.283203125, | |
| "learning_rate": 1.794871794871795e-05, | |
| "loss": 1.7675, | |
| "step": 242 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.26953125, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 1.7799, | |
| "step": 243 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.251953125, | |
| "learning_rate": 1.5384615384615387e-05, | |
| "loss": 1.756, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.2392578125, | |
| "learning_rate": 1.4102564102564104e-05, | |
| "loss": 1.8738, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.265625, | |
| "learning_rate": 1.282051282051282e-05, | |
| "loss": 1.7766, | |
| "step": 246 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.25, | |
| "learning_rate": 1.153846153846154e-05, | |
| "loss": 1.8446, | |
| "step": 247 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.265625, | |
| "learning_rate": 1.0256410256410256e-05, | |
| "loss": 1.885, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.29296875, | |
| "learning_rate": 8.974358974358976e-06, | |
| "loss": 1.809, | |
| "step": 249 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.265625, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 1.7866, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "eval_loss": 1.810941219329834, | |
| "eval_runtime": 125.5076, | |
| "eval_samples_per_second": 39.838, | |
| "eval_steps_per_second": 1.251, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.263671875, | |
| "learning_rate": 6.41025641025641e-06, | |
| "loss": 1.8828, | |
| "step": 251 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.2890625, | |
| "learning_rate": 5.128205128205128e-06, | |
| "loss": 1.8189, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.25, | |
| "learning_rate": 3.846153846153847e-06, | |
| "loss": 1.7144, | |
| "step": 253 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 0.2734375, | |
| "learning_rate": 2.564102564102564e-06, | |
| "loss": 1.7287, | |
| "step": 254 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 0.251953125, | |
| "learning_rate": 1.282051282051282e-06, | |
| "loss": 1.8007, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 0.263671875, | |
| "learning_rate": 0.0, | |
| "loss": 1.8333, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "step": 256, | |
| "total_flos": 1.8626901413206426e+17, | |
| "train_loss": 2.2480679317377508, | |
| "train_runtime": 4476.2307, | |
| "train_samples_per_second": 3.66, | |
| "train_steps_per_second": 0.057 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eval_loss": 1.810941219329834, | |
| "eval_runtime": 125.8181, | |
| "eval_samples_per_second": 39.74, | |
| "eval_steps_per_second": 1.248, | |
| "step": 256 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 256, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 10, | |
| "total_flos": 1.8626901413206426e+17, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |