| { | |
| "best_metric": 1.5016273260116577, | |
| "best_model_checkpoint": "lora_lr/mistralai/Mistral-7B-Instruct-v0.2/unaligned/checkpoint-240", | |
| "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": 1.484375, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 3.7421, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 1.4921875, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 3.6136, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 1.4140625, | |
| "learning_rate": 6e-06, | |
| "loss": 3.5641, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 1.59375, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 3.8168, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 1.5703125, | |
| "learning_rate": 1e-05, | |
| "loss": 3.7471, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 1.4609375, | |
| "learning_rate": 1.2e-05, | |
| "loss": 3.6939, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 1.5546875, | |
| "learning_rate": 1.4000000000000001e-05, | |
| "loss": 3.5964, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 1.5625, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 3.7708, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 1.5703125, | |
| "learning_rate": 1.8e-05, | |
| "loss": 3.7418, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 1.609375, | |
| "learning_rate": 2e-05, | |
| "loss": 3.8394, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "eval_loss": 3.672560214996338, | |
| "eval_runtime": 108.3497, | |
| "eval_samples_per_second": 46.147, | |
| "eval_steps_per_second": 1.449, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 1.5078125, | |
| "learning_rate": 2.2000000000000003e-05, | |
| "loss": 3.6075, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 1.4296875, | |
| "learning_rate": 2.4e-05, | |
| "loss": 3.4171, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 1.6484375, | |
| "learning_rate": 2.6000000000000002e-05, | |
| "loss": 3.7848, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 1.6328125, | |
| "learning_rate": 2.8000000000000003e-05, | |
| "loss": 3.7797, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 1.609375, | |
| "learning_rate": 3e-05, | |
| "loss": 3.7132, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 1.5390625, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 3.5353, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 1.828125, | |
| "learning_rate": 3.4000000000000007e-05, | |
| "loss": 3.687, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 1.65625, | |
| "learning_rate": 3.6e-05, | |
| "loss": 3.4312, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 1.8203125, | |
| "learning_rate": 3.8e-05, | |
| "loss": 3.4971, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 1.875, | |
| "learning_rate": 4e-05, | |
| "loss": 3.3386, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "eval_loss": 3.376652240753174, | |
| "eval_runtime": 108.6461, | |
| "eval_samples_per_second": 46.021, | |
| "eval_steps_per_second": 1.445, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 1.9921875, | |
| "learning_rate": 4.2e-05, | |
| "loss": 3.3568, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 2.109375, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 3.3976, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 2.125, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 3.3644, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 2.28125, | |
| "learning_rate": 4.8e-05, | |
| "loss": 3.1825, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 2.421875, | |
| "learning_rate": 5e-05, | |
| "loss": 3.2047, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 2.359375, | |
| "learning_rate": 5.2000000000000004e-05, | |
| "loss": 3.098, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 2.390625, | |
| "learning_rate": 5.4000000000000005e-05, | |
| "loss": 2.9831, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 2.359375, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 2.9573, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 2.609375, | |
| "learning_rate": 5.8e-05, | |
| "loss": 2.9497, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 2.125, | |
| "learning_rate": 6e-05, | |
| "loss": 2.5156, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "eval_loss": 2.578493356704712, | |
| "eval_runtime": 109.4296, | |
| "eval_samples_per_second": 45.691, | |
| "eval_steps_per_second": 1.435, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 2.25, | |
| "learning_rate": 6.2e-05, | |
| "loss": 2.6802, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 1.984375, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 2.4563, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 2.046875, | |
| "learning_rate": 6.6e-05, | |
| "loss": 2.4709, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 2.046875, | |
| "learning_rate": 6.800000000000001e-05, | |
| "loss": 2.4096, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 1.75, | |
| "learning_rate": 7e-05, | |
| "loss": 2.2541, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 1.78125, | |
| "learning_rate": 7.2e-05, | |
| "loss": 2.1879, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 1.6484375, | |
| "learning_rate": 7.4e-05, | |
| "loss": 2.2291, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 1.4140625, | |
| "learning_rate": 7.6e-05, | |
| "loss": 2.028, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 1.1640625, | |
| "learning_rate": 7.800000000000001e-05, | |
| "loss": 1.9569, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.9296875, | |
| "learning_rate": 8e-05, | |
| "loss": 1.8834, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "eval_loss": 1.834492564201355, | |
| "eval_runtime": 109.4942, | |
| "eval_samples_per_second": 45.665, | |
| "eval_steps_per_second": 1.434, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.765625, | |
| "learning_rate": 8.2e-05, | |
| "loss": 1.9118, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.640625, | |
| "learning_rate": 8.4e-05, | |
| "loss": 1.8096, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.515625, | |
| "learning_rate": 8.6e-05, | |
| "loss": 1.8097, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.49609375, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 1.7625, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.578125, | |
| "learning_rate": 9e-05, | |
| "loss": 1.7719, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.490234375, | |
| "learning_rate": 9.200000000000001e-05, | |
| "loss": 1.7783, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.5625, | |
| "learning_rate": 9.4e-05, | |
| "loss": 1.7888, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.55078125, | |
| "learning_rate": 9.6e-05, | |
| "loss": 1.7738, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.5859375, | |
| "learning_rate": 9.8e-05, | |
| "loss": 1.7428, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.55078125, | |
| "learning_rate": 0.0001, | |
| "loss": 1.7651, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "eval_loss": 1.744554042816162, | |
| "eval_runtime": 109.7416, | |
| "eval_samples_per_second": 45.562, | |
| "eval_steps_per_second": 1.431, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.58984375, | |
| "learning_rate": 0.00010200000000000001, | |
| "loss": 1.8558, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.6171875, | |
| "learning_rate": 0.00010400000000000001, | |
| "loss": 1.7136, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.6328125, | |
| "learning_rate": 0.00010600000000000002, | |
| "loss": 1.6953, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.79296875, | |
| "learning_rate": 0.00010800000000000001, | |
| "loss": 1.7225, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.83203125, | |
| "learning_rate": 0.00011000000000000002, | |
| "loss": 1.6275, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.84375, | |
| "learning_rate": 0.00011200000000000001, | |
| "loss": 1.6724, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.5546875, | |
| "learning_rate": 0.00011399999999999999, | |
| "loss": 1.627, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.388671875, | |
| "learning_rate": 0.000116, | |
| "loss": 1.6624, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.34375, | |
| "learning_rate": 0.000118, | |
| "loss": 1.592, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.306640625, | |
| "learning_rate": 0.00012, | |
| "loss": 1.638, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "eval_loss": 1.5958868265151978, | |
| "eval_runtime": 109.6677, | |
| "eval_samples_per_second": 45.592, | |
| "eval_steps_per_second": 1.432, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.302734375, | |
| "learning_rate": 0.000122, | |
| "loss": 1.5942, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.30078125, | |
| "learning_rate": 0.000124, | |
| "loss": 1.6169, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.296875, | |
| "learning_rate": 0.000126, | |
| "loss": 1.6183, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.3046875, | |
| "learning_rate": 0.00012800000000000002, | |
| "loss": 1.5629, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.3203125, | |
| "learning_rate": 0.00013000000000000002, | |
| "loss": 1.6236, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.267578125, | |
| "learning_rate": 0.000132, | |
| "loss": 1.6701, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.294921875, | |
| "learning_rate": 0.000134, | |
| "loss": 1.6067, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.30859375, | |
| "learning_rate": 0.00013600000000000003, | |
| "loss": 1.5935, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.27734375, | |
| "learning_rate": 0.000138, | |
| "loss": 1.578, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.26953125, | |
| "learning_rate": 0.00014, | |
| "loss": 1.592, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "eval_loss": 1.571932077407837, | |
| "eval_runtime": 109.6303, | |
| "eval_samples_per_second": 45.608, | |
| "eval_steps_per_second": 1.432, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.28125, | |
| "learning_rate": 0.000142, | |
| "loss": 1.5923, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.283203125, | |
| "learning_rate": 0.000144, | |
| "loss": 1.5838, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.26171875, | |
| "learning_rate": 0.000146, | |
| "loss": 1.5646, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.255859375, | |
| "learning_rate": 0.000148, | |
| "loss": 1.6265, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.271484375, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 1.622, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.2373046875, | |
| "learning_rate": 0.000152, | |
| "loss": 1.5504, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.267578125, | |
| "learning_rate": 0.000154, | |
| "loss": 1.6184, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.228515625, | |
| "learning_rate": 0.00015600000000000002, | |
| "loss": 1.5829, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.2451171875, | |
| "learning_rate": 0.00015800000000000002, | |
| "loss": 1.5305, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.25, | |
| "learning_rate": 0.00016, | |
| "loss": 1.554, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_loss": 1.5540077686309814, | |
| "eval_runtime": 109.5859, | |
| "eval_samples_per_second": 45.626, | |
| "eval_steps_per_second": 1.433, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.24609375, | |
| "learning_rate": 0.000162, | |
| "loss": 1.5691, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.2470703125, | |
| "learning_rate": 0.000164, | |
| "loss": 1.542, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.2431640625, | |
| "learning_rate": 0.000166, | |
| "loss": 1.5844, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.2333984375, | |
| "learning_rate": 0.000168, | |
| "loss": 1.5429, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.2255859375, | |
| "learning_rate": 0.00017, | |
| "loss": 1.5836, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.25390625, | |
| "learning_rate": 0.000172, | |
| "loss": 1.5227, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.2412109375, | |
| "learning_rate": 0.000174, | |
| "loss": 1.5495, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.21484375, | |
| "learning_rate": 0.00017600000000000002, | |
| "loss": 1.6124, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.2470703125, | |
| "learning_rate": 0.00017800000000000002, | |
| "loss": 1.5469, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.2470703125, | |
| "learning_rate": 0.00018, | |
| "loss": 1.5215, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "eval_loss": 1.5420269966125488, | |
| "eval_runtime": 109.6203, | |
| "eval_samples_per_second": 45.612, | |
| "eval_steps_per_second": 1.432, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.22265625, | |
| "learning_rate": 0.000182, | |
| "loss": 1.5654, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.2392578125, | |
| "learning_rate": 0.00018400000000000003, | |
| "loss": 1.494, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.216796875, | |
| "learning_rate": 0.00018600000000000002, | |
| "loss": 1.6285, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.2451171875, | |
| "learning_rate": 0.000188, | |
| "loss": 1.5177, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.2177734375, | |
| "learning_rate": 0.00019, | |
| "loss": 1.5659, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.220703125, | |
| "learning_rate": 0.000192, | |
| "loss": 1.6005, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.251953125, | |
| "learning_rate": 0.000194, | |
| "loss": 1.5713, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 0.2294921875, | |
| "learning_rate": 0.000196, | |
| "loss": 1.4815, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 0.216796875, | |
| "learning_rate": 0.00019800000000000002, | |
| "loss": 1.489, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 0.216796875, | |
| "learning_rate": 0.0002, | |
| "loss": 1.497, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "eval_loss": 1.5331306457519531, | |
| "eval_runtime": 108.6942, | |
| "eval_samples_per_second": 46.001, | |
| "eval_steps_per_second": 1.444, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 0.23046875, | |
| "learning_rate": 0.00019871794871794874, | |
| "loss": 1.4933, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 0.23828125, | |
| "learning_rate": 0.00019743589743589744, | |
| "loss": 1.5385, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 0.2431640625, | |
| "learning_rate": 0.00019615384615384615, | |
| "loss": 1.5745, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 0.220703125, | |
| "learning_rate": 0.00019487179487179487, | |
| "loss": 1.5485, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 0.248046875, | |
| "learning_rate": 0.0001935897435897436, | |
| "loss": 1.5391, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.23046875, | |
| "learning_rate": 0.00019230769230769233, | |
| "loss": 1.5649, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.212890625, | |
| "learning_rate": 0.00019102564102564104, | |
| "loss": 1.5543, | |
| "step": 107 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.23046875, | |
| "learning_rate": 0.00018974358974358974, | |
| "loss": 1.5377, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.2314453125, | |
| "learning_rate": 0.00018846153846153847, | |
| "loss": 1.534, | |
| "step": 109 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.2119140625, | |
| "learning_rate": 0.0001871794871794872, | |
| "loss": 1.5583, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "eval_loss": 1.526591181755066, | |
| "eval_runtime": 108.4319, | |
| "eval_samples_per_second": 46.112, | |
| "eval_steps_per_second": 1.448, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.234375, | |
| "learning_rate": 0.0001858974358974359, | |
| "loss": 1.4938, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.248046875, | |
| "learning_rate": 0.00018461538461538463, | |
| "loss": 1.5327, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.2294921875, | |
| "learning_rate": 0.00018333333333333334, | |
| "loss": 1.5475, | |
| "step": 113 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.2431640625, | |
| "learning_rate": 0.00018205128205128207, | |
| "loss": 1.5718, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.23046875, | |
| "learning_rate": 0.00018076923076923077, | |
| "loss": 1.53, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.2177734375, | |
| "learning_rate": 0.0001794871794871795, | |
| "loss": 1.5675, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.2431640625, | |
| "learning_rate": 0.00017820512820512823, | |
| "loss": 1.6159, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.2119140625, | |
| "learning_rate": 0.00017692307692307693, | |
| "loss": 1.5419, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.2294921875, | |
| "learning_rate": 0.00017564102564102566, | |
| "loss": 1.4955, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.2236328125, | |
| "learning_rate": 0.00017435897435897436, | |
| "loss": 1.5639, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 1.521650791168213, | |
| "eval_runtime": 108.8276, | |
| "eval_samples_per_second": 45.944, | |
| "eval_steps_per_second": 1.443, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.2119140625, | |
| "learning_rate": 0.0001730769230769231, | |
| "loss": 1.5569, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.24609375, | |
| "learning_rate": 0.0001717948717948718, | |
| "loss": 1.527, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.2197265625, | |
| "learning_rate": 0.00017051282051282053, | |
| "loss": 1.538, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.26953125, | |
| "learning_rate": 0.00016923076923076923, | |
| "loss": 1.5353, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.2353515625, | |
| "learning_rate": 0.00016794871794871796, | |
| "loss": 1.5495, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.2421875, | |
| "learning_rate": 0.0001666666666666667, | |
| "loss": 1.5366, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.228515625, | |
| "learning_rate": 0.0001653846153846154, | |
| "loss": 1.5695, | |
| "step": 127 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.2314453125, | |
| "learning_rate": 0.0001641025641025641, | |
| "loss": 1.5112, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.234375, | |
| "learning_rate": 0.00016282051282051282, | |
| "loss": 1.5919, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.2216796875, | |
| "learning_rate": 0.00016153846153846155, | |
| "loss": 1.5785, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "eval_loss": 1.517764925956726, | |
| "eval_runtime": 109.0406, | |
| "eval_samples_per_second": 45.854, | |
| "eval_steps_per_second": 1.44, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.248046875, | |
| "learning_rate": 0.00016025641025641028, | |
| "loss": 1.5056, | |
| "step": 131 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.236328125, | |
| "learning_rate": 0.00015897435897435896, | |
| "loss": 1.5427, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.23046875, | |
| "learning_rate": 0.0001576923076923077, | |
| "loss": 1.5357, | |
| "step": 133 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.21875, | |
| "learning_rate": 0.00015641025641025642, | |
| "loss": 1.5671, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.2421875, | |
| "learning_rate": 0.00015512820512820515, | |
| "loss": 1.5247, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.2412109375, | |
| "learning_rate": 0.00015384615384615385, | |
| "loss": 1.557, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.26953125, | |
| "learning_rate": 0.00015256410256410255, | |
| "loss": 1.5042, | |
| "step": 137 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.2373046875, | |
| "learning_rate": 0.00015128205128205128, | |
| "loss": 1.5061, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.26953125, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 1.5608, | |
| "step": 139 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.25, | |
| "learning_rate": 0.00014871794871794872, | |
| "loss": 1.5235, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "eval_loss": 1.5145833492279053, | |
| "eval_runtime": 109.0187, | |
| "eval_samples_per_second": 45.864, | |
| "eval_steps_per_second": 1.44, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.2314453125, | |
| "learning_rate": 0.00014743589743589745, | |
| "loss": 1.5099, | |
| "step": 141 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.228515625, | |
| "learning_rate": 0.00014615384615384615, | |
| "loss": 1.5413, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.2373046875, | |
| "learning_rate": 0.00014487179487179488, | |
| "loss": 1.5383, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.234375, | |
| "learning_rate": 0.0001435897435897436, | |
| "loss": 1.5671, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.244140625, | |
| "learning_rate": 0.0001423076923076923, | |
| "loss": 1.5102, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.220703125, | |
| "learning_rate": 0.00014102564102564104, | |
| "loss": 1.545, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.228515625, | |
| "learning_rate": 0.00013974358974358974, | |
| "loss": 1.5223, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.2275390625, | |
| "learning_rate": 0.00013846153846153847, | |
| "loss": 1.433, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.2353515625, | |
| "learning_rate": 0.00013717948717948718, | |
| "loss": 1.554, | |
| "step": 149 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.2314453125, | |
| "learning_rate": 0.0001358974358974359, | |
| "loss": 1.5447, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "eval_loss": 1.5118788480758667, | |
| "eval_runtime": 109.1366, | |
| "eval_samples_per_second": 45.814, | |
| "eval_steps_per_second": 1.439, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.259765625, | |
| "learning_rate": 0.00013461538461538464, | |
| "loss": 1.4771, | |
| "step": 151 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.240234375, | |
| "learning_rate": 0.00013333333333333334, | |
| "loss": 1.5853, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.2197265625, | |
| "learning_rate": 0.00013205128205128204, | |
| "loss": 1.5237, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.2578125, | |
| "learning_rate": 0.00013076923076923077, | |
| "loss": 1.4963, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.2373046875, | |
| "learning_rate": 0.0001294871794871795, | |
| "loss": 1.5393, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.220703125, | |
| "learning_rate": 0.00012820512820512823, | |
| "loss": 1.4725, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.220703125, | |
| "learning_rate": 0.00012692307692307693, | |
| "loss": 1.5187, | |
| "step": 157 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.251953125, | |
| "learning_rate": 0.00012564102564102564, | |
| "loss": 1.5263, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.2470703125, | |
| "learning_rate": 0.00012435897435897437, | |
| "loss": 1.5388, | |
| "step": 159 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.224609375, | |
| "learning_rate": 0.0001230769230769231, | |
| "loss": 1.4606, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_loss": 1.5093975067138672, | |
| "eval_runtime": 109.3522, | |
| "eval_samples_per_second": 45.724, | |
| "eval_steps_per_second": 1.436, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.2392578125, | |
| "learning_rate": 0.00012179487179487179, | |
| "loss": 1.5584, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.224609375, | |
| "learning_rate": 0.00012051282051282052, | |
| "loss": 1.4941, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.236328125, | |
| "learning_rate": 0.00011923076923076923, | |
| "loss": 1.5283, | |
| "step": 163 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.2373046875, | |
| "learning_rate": 0.00011794871794871796, | |
| "loss": 1.4864, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.24609375, | |
| "learning_rate": 0.00011666666666666668, | |
| "loss": 1.4398, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.251953125, | |
| "learning_rate": 0.00011538461538461538, | |
| "loss": 1.514, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.25390625, | |
| "learning_rate": 0.0001141025641025641, | |
| "loss": 1.5578, | |
| "step": 167 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.2353515625, | |
| "learning_rate": 0.00011282051282051283, | |
| "loss": 1.5682, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.23046875, | |
| "learning_rate": 0.00011153846153846154, | |
| "loss": 1.4911, | |
| "step": 169 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.236328125, | |
| "learning_rate": 0.00011025641025641027, | |
| "loss": 1.5906, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_loss": 1.5074635744094849, | |
| "eval_runtime": 108.9797, | |
| "eval_samples_per_second": 45.88, | |
| "eval_steps_per_second": 1.441, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.271484375, | |
| "learning_rate": 0.00010897435897435896, | |
| "loss": 1.5266, | |
| "step": 171 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.23828125, | |
| "learning_rate": 0.0001076923076923077, | |
| "loss": 1.4515, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.2451171875, | |
| "learning_rate": 0.00010641025641025641, | |
| "loss": 1.5492, | |
| "step": 173 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.2314453125, | |
| "learning_rate": 0.00010512820512820514, | |
| "loss": 1.4961, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.2421875, | |
| "learning_rate": 0.00010384615384615386, | |
| "loss": 1.5594, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.2275390625, | |
| "learning_rate": 0.00010256410256410256, | |
| "loss": 1.5469, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.240234375, | |
| "learning_rate": 0.00010128205128205129, | |
| "loss": 1.4925, | |
| "step": 177 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.2578125, | |
| "learning_rate": 0.0001, | |
| "loss": 1.4808, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.251953125, | |
| "learning_rate": 9.871794871794872e-05, | |
| "loss": 1.4963, | |
| "step": 179 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.2578125, | |
| "learning_rate": 9.743589743589744e-05, | |
| "loss": 1.5397, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "eval_loss": 1.5060302019119263, | |
| "eval_runtime": 109.1672, | |
| "eval_samples_per_second": 45.801, | |
| "eval_steps_per_second": 1.438, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.2451171875, | |
| "learning_rate": 9.615384615384617e-05, | |
| "loss": 1.5202, | |
| "step": 181 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.2158203125, | |
| "learning_rate": 9.487179487179487e-05, | |
| "loss": 1.488, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.25390625, | |
| "learning_rate": 9.35897435897436e-05, | |
| "loss": 1.5124, | |
| "step": 183 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.2578125, | |
| "learning_rate": 9.230769230769232e-05, | |
| "loss": 1.4791, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.2421875, | |
| "learning_rate": 9.102564102564103e-05, | |
| "loss": 1.6039, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.2353515625, | |
| "learning_rate": 8.974358974358975e-05, | |
| "loss": 1.5389, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.2216796875, | |
| "learning_rate": 8.846153846153847e-05, | |
| "loss": 1.5542, | |
| "step": 187 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.2353515625, | |
| "learning_rate": 8.717948717948718e-05, | |
| "loss": 1.5131, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.248046875, | |
| "learning_rate": 8.58974358974359e-05, | |
| "loss": 1.5371, | |
| "step": 189 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.2392578125, | |
| "learning_rate": 8.461538461538461e-05, | |
| "loss": 1.5264, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_loss": 1.5048484802246094, | |
| "eval_runtime": 109.1034, | |
| "eval_samples_per_second": 45.828, | |
| "eval_steps_per_second": 1.439, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.236328125, | |
| "learning_rate": 8.333333333333334e-05, | |
| "loss": 1.5295, | |
| "step": 191 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.2333984375, | |
| "learning_rate": 8.205128205128205e-05, | |
| "loss": 1.5677, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.255859375, | |
| "learning_rate": 8.076923076923078e-05, | |
| "loss": 1.4473, | |
| "step": 193 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.2353515625, | |
| "learning_rate": 7.948717948717948e-05, | |
| "loss": 1.5756, | |
| "step": 194 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.236328125, | |
| "learning_rate": 7.820512820512821e-05, | |
| "loss": 1.4832, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.2294921875, | |
| "learning_rate": 7.692307692307693e-05, | |
| "loss": 1.5144, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.248046875, | |
| "learning_rate": 7.564102564102564e-05, | |
| "loss": 1.5077, | |
| "step": 197 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.21875, | |
| "learning_rate": 7.435897435897436e-05, | |
| "loss": 1.5042, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.22265625, | |
| "learning_rate": 7.307692307692307e-05, | |
| "loss": 1.5022, | |
| "step": 199 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.265625, | |
| "learning_rate": 7.17948717948718e-05, | |
| "loss": 1.484, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "eval_loss": 1.503655195236206, | |
| "eval_runtime": 117.4418, | |
| "eval_samples_per_second": 42.574, | |
| "eval_steps_per_second": 1.337, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.2431640625, | |
| "learning_rate": 7.051282051282052e-05, | |
| "loss": 1.4973, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.2314453125, | |
| "learning_rate": 6.923076923076924e-05, | |
| "loss": 1.5267, | |
| "step": 202 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.2236328125, | |
| "learning_rate": 6.794871794871795e-05, | |
| "loss": 1.5657, | |
| "step": 203 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.23046875, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 1.5184, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.23828125, | |
| "learning_rate": 6.538461538461539e-05, | |
| "loss": 1.4522, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.2451171875, | |
| "learning_rate": 6.410256410256412e-05, | |
| "loss": 1.5309, | |
| "step": 206 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.2373046875, | |
| "learning_rate": 6.282051282051282e-05, | |
| "loss": 1.4503, | |
| "step": 207 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.2265625, | |
| "learning_rate": 6.153846153846155e-05, | |
| "loss": 1.5173, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.275390625, | |
| "learning_rate": 6.025641025641026e-05, | |
| "loss": 1.5112, | |
| "step": 209 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.234375, | |
| "learning_rate": 5.897435897435898e-05, | |
| "loss": 1.5342, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "eval_loss": 1.5029850006103516, | |
| "eval_runtime": 118.7123, | |
| "eval_samples_per_second": 42.119, | |
| "eval_steps_per_second": 1.323, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.2333984375, | |
| "learning_rate": 5.769230769230769e-05, | |
| "loss": 1.5216, | |
| "step": 211 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.2373046875, | |
| "learning_rate": 5.6410256410256414e-05, | |
| "loss": 1.5005, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.23046875, | |
| "learning_rate": 5.512820512820514e-05, | |
| "loss": 1.5372, | |
| "step": 213 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.2353515625, | |
| "learning_rate": 5.384615384615385e-05, | |
| "loss": 1.4916, | |
| "step": 214 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.232421875, | |
| "learning_rate": 5.256410256410257e-05, | |
| "loss": 1.4511, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.251953125, | |
| "learning_rate": 5.128205128205128e-05, | |
| "loss": 1.4984, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.275390625, | |
| "learning_rate": 5e-05, | |
| "loss": 1.5332, | |
| "step": 217 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.2412109375, | |
| "learning_rate": 4.871794871794872e-05, | |
| "loss": 1.558, | |
| "step": 218 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.255859375, | |
| "learning_rate": 4.7435897435897435e-05, | |
| "loss": 1.5057, | |
| "step": 219 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.2431640625, | |
| "learning_rate": 4.615384615384616e-05, | |
| "loss": 1.5374, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "eval_loss": 1.5021294355392456, | |
| "eval_runtime": 120.5989, | |
| "eval_samples_per_second": 41.46, | |
| "eval_steps_per_second": 1.302, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.26953125, | |
| "learning_rate": 4.4871794871794874e-05, | |
| "loss": 1.5332, | |
| "step": 221 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.236328125, | |
| "learning_rate": 4.358974358974359e-05, | |
| "loss": 1.5205, | |
| "step": 222 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.224609375, | |
| "learning_rate": 4.230769230769231e-05, | |
| "loss": 1.497, | |
| "step": 223 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.22265625, | |
| "learning_rate": 4.1025641025641023e-05, | |
| "loss": 1.5158, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.2421875, | |
| "learning_rate": 3.974358974358974e-05, | |
| "loss": 1.4847, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.244140625, | |
| "learning_rate": 3.846153846153846e-05, | |
| "loss": 1.4456, | |
| "step": 226 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.228515625, | |
| "learning_rate": 3.717948717948718e-05, | |
| "loss": 1.5334, | |
| "step": 227 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.25390625, | |
| "learning_rate": 3.58974358974359e-05, | |
| "loss": 1.5238, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.2490234375, | |
| "learning_rate": 3.461538461538462e-05, | |
| "loss": 1.4388, | |
| "step": 229 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.2353515625, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 1.5555, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": 1.5019102096557617, | |
| "eval_runtime": 109.1608, | |
| "eval_samples_per_second": 45.804, | |
| "eval_steps_per_second": 1.438, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.267578125, | |
| "learning_rate": 3.205128205128206e-05, | |
| "loss": 1.5535, | |
| "step": 231 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.240234375, | |
| "learning_rate": 3.0769230769230774e-05, | |
| "loss": 1.4322, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.2470703125, | |
| "learning_rate": 2.948717948717949e-05, | |
| "loss": 1.471, | |
| "step": 233 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.25, | |
| "learning_rate": 2.8205128205128207e-05, | |
| "loss": 1.5444, | |
| "step": 234 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.25, | |
| "learning_rate": 2.6923076923076923e-05, | |
| "loss": 1.5331, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.236328125, | |
| "learning_rate": 2.564102564102564e-05, | |
| "loss": 1.5355, | |
| "step": 236 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.244140625, | |
| "learning_rate": 2.435897435897436e-05, | |
| "loss": 1.4883, | |
| "step": 237 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.263671875, | |
| "learning_rate": 2.307692307692308e-05, | |
| "loss": 1.5266, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.2451171875, | |
| "learning_rate": 2.1794871794871795e-05, | |
| "loss": 1.5576, | |
| "step": 239 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.259765625, | |
| "learning_rate": 2.0512820512820512e-05, | |
| "loss": 1.5615, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eval_loss": 1.5016273260116577, | |
| "eval_runtime": 109.191, | |
| "eval_samples_per_second": 45.791, | |
| "eval_steps_per_second": 1.438, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.23828125, | |
| "learning_rate": 1.923076923076923e-05, | |
| "loss": 1.531, | |
| "step": 241 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.2470703125, | |
| "learning_rate": 1.794871794871795e-05, | |
| "loss": 1.4563, | |
| "step": 242 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.24609375, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 1.4773, | |
| "step": 243 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.2431640625, | |
| "learning_rate": 1.5384615384615387e-05, | |
| "loss": 1.4496, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.228515625, | |
| "learning_rate": 1.4102564102564104e-05, | |
| "loss": 1.5677, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.2578125, | |
| "learning_rate": 1.282051282051282e-05, | |
| "loss": 1.4935, | |
| "step": 246 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.2216796875, | |
| "learning_rate": 1.153846153846154e-05, | |
| "loss": 1.5357, | |
| "step": 247 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.259765625, | |
| "learning_rate": 1.0256410256410256e-05, | |
| "loss": 1.5693, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.271484375, | |
| "learning_rate": 8.974358974358976e-06, | |
| "loss": 1.5037, | |
| "step": 249 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.236328125, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 1.4786, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "eval_loss": 1.501659631729126, | |
| "eval_runtime": 109.1682, | |
| "eval_samples_per_second": 45.801, | |
| "eval_steps_per_second": 1.438, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.2412109375, | |
| "learning_rate": 6.41025641025641e-06, | |
| "loss": 1.5624, | |
| "step": 251 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.232421875, | |
| "learning_rate": 5.128205128205128e-06, | |
| "loss": 1.4851, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.2451171875, | |
| "learning_rate": 3.846153846153847e-06, | |
| "loss": 1.4184, | |
| "step": 253 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 0.2353515625, | |
| "learning_rate": 2.564102564102564e-06, | |
| "loss": 1.4701, | |
| "step": 254 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 0.244140625, | |
| "learning_rate": 1.282051282051282e-06, | |
| "loss": 1.5079, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 0.2216796875, | |
| "learning_rate": 0.0, | |
| "loss": 1.5386, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "step": 256, | |
| "total_flos": 1.8280284932328653e+17, | |
| "train_loss": 1.802713230252266, | |
| "train_runtime": 3920.872, | |
| "train_samples_per_second": 4.179, | |
| "train_steps_per_second": 0.065 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eval_loss": 1.5016273260116577, | |
| "eval_runtime": 109.2744, | |
| "eval_samples_per_second": 45.756, | |
| "eval_steps_per_second": 1.437, | |
| "step": 256 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 256, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 10, | |
| "total_flos": 1.8280284932328653e+17, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |