Llama2_Instruction_Finetuning_Experiments
/
llama2_7b_chat_adapter
/checkpoint-350
/trainer_state.json
| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.865533230293663, | |
| "eval_steps": 500, | |
| "global_step": 350, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 0.0003, | |
| "loss": 1.8879, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 0.0003, | |
| "loss": 1.7265, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.0003, | |
| "loss": 1.715, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.0003, | |
| "loss": 1.6706, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.0003, | |
| "loss": 1.4835, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.0003, | |
| "loss": 1.4257, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.0003, | |
| "loss": 1.3796, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.0003, | |
| "loss": 1.3675, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.0003, | |
| "loss": 1.312, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.0003, | |
| "loss": 1.1967, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.0003, | |
| "loss": 1.1773, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.0003, | |
| "loss": 1.1163, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.0003, | |
| "loss": 1.0598, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.0003, | |
| "loss": 1.0714, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.0003, | |
| "loss": 1.0266, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.0003, | |
| "loss": 0.9998, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.0003, | |
| "loss": 0.9691, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.0003, | |
| "loss": 0.9833, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.0003, | |
| "loss": 1.009, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.0003, | |
| "loss": 1.0057, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.0003, | |
| "loss": 0.9791, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.0003, | |
| "loss": 1.0029, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.0003, | |
| "loss": 1.0009, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.0003, | |
| "loss": 0.9382, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.0003, | |
| "loss": 0.9214, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.0003, | |
| "loss": 0.9107, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0003, | |
| "loss": 0.9951, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0003, | |
| "loss": 0.9181, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0003, | |
| "loss": 0.9082, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8779, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8996, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.0003, | |
| "loss": 0.9196, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8763, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.0003, | |
| "loss": 0.9125, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8474, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8901, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8782, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.0003, | |
| "loss": 0.863, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8747, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8428, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8868, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8729, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8606, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8643, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8372, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8853, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8399, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8536, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.0003, | |
| "loss": 0.857, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8858, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8354, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8429, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8442, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8433, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8594, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.0003, | |
| "loss": 0.857, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8918, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8335, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8188, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8483, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8442, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0003, | |
| "loss": 0.855, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8465, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8503, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8588, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8444, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8587, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8951, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8274, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.0003, | |
| "loss": 0.857, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8602, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7978, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8499, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8525, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8534, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8437, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8468, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8468, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0003, | |
| "loss": 0.84, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0003, | |
| "loss": 0.832, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8038, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8481, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8464, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8361, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8262, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8579, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8397, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8508, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8485, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8435, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8298, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8577, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8658, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8329, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8175, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8371, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8386, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8171, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8672, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8458, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8553, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7931, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8094, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8471, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8241, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8516, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8457, | |
| "step": 107 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8505, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.0003, | |
| "loss": 0.813, | |
| "step": 109 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.0003, | |
| "loss": 0.829, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8475, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8582, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8405, | |
| "step": 113 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8031, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7739, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8314, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0003, | |
| "loss": 0.786, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8348, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8297, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.0003, | |
| "loss": 0.841, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8344, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8321, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.0003, | |
| "loss": 0.839, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8254, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0003, | |
| "loss": 0.85, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0003, | |
| "loss": 0.863, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8202, | |
| "step": 127 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8325, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.0003, | |
| "loss": 0.819, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8182, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8917, | |
| "step": 131 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.0003, | |
| "loss": 0.871, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7837, | |
| "step": 133 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8002, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8526, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8311, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8, | |
| "step": 137 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8615, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8225, | |
| "step": 139 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8297, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8379, | |
| "step": 141 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8539, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8026, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8013, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8244, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8544, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8111, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8497, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8004, | |
| "step": 149 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8349, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8079, | |
| "step": 151 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8424, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8272, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8476, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8296, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8416, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7948, | |
| "step": 157 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8095, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8302, | |
| "step": 159 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8077, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8315, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0003, | |
| "loss": 0.883, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8057, | |
| "step": 163 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8139, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.0003, | |
| "loss": 0.796, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8329, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8513, | |
| "step": 167 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8637, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8467, | |
| "step": 169 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8215, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8472, | |
| "step": 171 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.0003, | |
| "loss": 0.799, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8426, | |
| "step": 173 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8331, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.0003, | |
| "loss": 0.827, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8272, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8526, | |
| "step": 177 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8539, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8292, | |
| "step": 179 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8262, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7878, | |
| "step": 181 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8161, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8459, | |
| "step": 183 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8613, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.0003, | |
| "loss": 0.854, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8475, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8303, | |
| "step": 187 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8283, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.0003, | |
| "loss": 0.86, | |
| "step": 189 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8531, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8026, | |
| "step": 191 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8341, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8463, | |
| "step": 193 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8455, | |
| "step": 194 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8361, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8225, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8445, | |
| "step": 197 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7892, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8264, | |
| "step": 199 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7936, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8242, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8065, | |
| "step": 202 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8286, | |
| "step": 203 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8499, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8524, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8309, | |
| "step": 206 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8667, | |
| "step": 207 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8281, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8145, | |
| "step": 209 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8177, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7898, | |
| "step": 211 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8347, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8248, | |
| "step": 213 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8796, | |
| "step": 214 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.0003, | |
| "loss": 0.824, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8425, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8123, | |
| "step": 217 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8169, | |
| "step": 218 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7943, | |
| "step": 219 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8175, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8607, | |
| "step": 221 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 0.0003, | |
| "loss": 0.833, | |
| "step": 222 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8655, | |
| "step": 223 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8231, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8317, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8154, | |
| "step": 226 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8321, | |
| "step": 227 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8035, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8446, | |
| "step": 229 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8307, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8435, | |
| "step": 231 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8364, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8314, | |
| "step": 233 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8231, | |
| "step": 234 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7987, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8145, | |
| "step": 236 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8392, | |
| "step": 237 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8521, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8341, | |
| "step": 239 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8697, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8091, | |
| "step": 241 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8177, | |
| "step": 242 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8296, | |
| "step": 243 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8139, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8274, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8213, | |
| "step": 246 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8455, | |
| "step": 247 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 0.0003, | |
| "loss": 0.821, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.0003, | |
| "loss": 0.817, | |
| "step": 249 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.0003, | |
| "loss": 0.77, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8415, | |
| "step": 251 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8359, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8382, | |
| "step": 253 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8315, | |
| "step": 254 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8237, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8373, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8052, | |
| "step": 257 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8056, | |
| "step": 258 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 0.0003, | |
| "loss": 0.835, | |
| "step": 259 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7921, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.0003, | |
| "loss": 0.801, | |
| "step": 261 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.0003, | |
| "loss": 0.877, | |
| "step": 262 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8166, | |
| "step": 263 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8402, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7769, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8147, | |
| "step": 266 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7838, | |
| "step": 267 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8153, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.0003, | |
| "loss": 0.837, | |
| "step": 269 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8599, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7842, | |
| "step": 271 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7927, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7914, | |
| "step": 273 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.0003, | |
| "loss": 0.827, | |
| "step": 274 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8264, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8216, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7953, | |
| "step": 277 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.0003, | |
| "loss": 0.832, | |
| "step": 278 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8411, | |
| "step": 279 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8176, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8215, | |
| "step": 281 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8164, | |
| "step": 282 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.0003, | |
| "loss": 0.81, | |
| "step": 283 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7927, | |
| "step": 284 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8206, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8109, | |
| "step": 286 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8128, | |
| "step": 287 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8425, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7934, | |
| "step": 289 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8115, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7676, | |
| "step": 291 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8037, | |
| "step": 292 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8143, | |
| "step": 293 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.0003, | |
| "loss": 0.812, | |
| "step": 294 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8221, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8055, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8349, | |
| "step": 297 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7962, | |
| "step": 298 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8279, | |
| "step": 299 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8228, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8061, | |
| "step": 301 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8013, | |
| "step": 302 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8364, | |
| "step": 303 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8027, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8108, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8377, | |
| "step": 306 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8152, | |
| "step": 307 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8394, | |
| "step": 308 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8446, | |
| "step": 309 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8337, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8132, | |
| "step": 311 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7772, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8186, | |
| "step": 313 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8363, | |
| "step": 314 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8073, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8252, | |
| "step": 316 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8545, | |
| "step": 317 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 0.0003, | |
| "loss": 0.81, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7744, | |
| "step": 319 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 0.0003, | |
| "loss": 0.774, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8365, | |
| "step": 321 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8398, | |
| "step": 322 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8507, | |
| "step": 323 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8123, | |
| "step": 324 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8159, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8171, | |
| "step": 326 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 0.0003, | |
| "loss": 0.828, | |
| "step": 327 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8101, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7998, | |
| "step": 329 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7998, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7956, | |
| "step": 331 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8522, | |
| "step": 332 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7982, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8289, | |
| "step": 334 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7944, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8348, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7888, | |
| "step": 337 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 0.0003, | |
| "loss": 0.818, | |
| "step": 338 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8129, | |
| "step": 339 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8423, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 0.0003, | |
| "loss": 0.807, | |
| "step": 341 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8189, | |
| "step": 342 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8385, | |
| "step": 343 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8364, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8172, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8416, | |
| "step": 346 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8185, | |
| "step": 347 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8236, | |
| "step": 348 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8098, | |
| "step": 349 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8105, | |
| "step": 350 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 404, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "total_flos": 3.9310521413861376e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |