| { | |
| "best_metric": 0.2015175074338913, | |
| "best_model_checkpoint": "./output_solor/exp_16/checkpoint-95", | |
| "epoch": 5.0, | |
| "eval_steps": 5, | |
| "global_step": 95, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 5.0000000000000004e-08, | |
| "loss": 1.3493, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "eval_loss": 1.2794537544250488, | |
| "eval_runtime": 20.3935, | |
| "eval_samples_per_second": 3.138, | |
| "eval_steps_per_second": 0.392, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 2.0000000000000002e-07, | |
| "loss": 1.301, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 4.5e-07, | |
| "loss": 1.2596, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 8.000000000000001e-07, | |
| "loss": 1.2522, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 1.2500000000000003e-06, | |
| "loss": 1.2483, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "eval_loss": 1.2768707275390625, | |
| "eval_runtime": 20.6266, | |
| "eval_samples_per_second": 3.103, | |
| "eval_steps_per_second": 0.388, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 1.8e-06, | |
| "loss": 1.3056, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 2.4500000000000003e-06, | |
| "loss": 1.4414, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 3.2000000000000003e-06, | |
| "loss": 1.2347, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 4.05e-06, | |
| "loss": 1.2399, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 5.000000000000001e-06, | |
| "loss": 1.2275, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "eval_loss": 1.2099063396453857, | |
| "eval_runtime": 20.6171, | |
| "eval_samples_per_second": 3.104, | |
| "eval_steps_per_second": 0.388, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 6.05e-06, | |
| "loss": 1.154, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 7.2e-06, | |
| "loss": 1.2325, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 8.45e-06, | |
| "loss": 1.1022, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 9.800000000000001e-06, | |
| "loss": 1.0295, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 1.1249999999999999e-05, | |
| "loss": 1.0529, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "eval_loss": 1.0723849534988403, | |
| "eval_runtime": 20.6213, | |
| "eval_samples_per_second": 3.104, | |
| "eval_steps_per_second": 0.388, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 1.2800000000000001e-05, | |
| "loss": 1.1019, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 1.4450000000000002e-05, | |
| "loss": 1.0158, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 1.62e-05, | |
| "loss": 1.0235, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 1.805e-05, | |
| "loss": 1.1272, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 2.0000000000000005e-05, | |
| "loss": 0.8642, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "eval_loss": 0.9709420204162598, | |
| "eval_runtime": 20.3773, | |
| "eval_samples_per_second": 3.141, | |
| "eval_steps_per_second": 0.393, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 2.2049999999999997e-05, | |
| "loss": 0.996, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 2.42e-05, | |
| "loss": 0.8983, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 2.6450000000000003e-05, | |
| "loss": 0.9876, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 2.88e-05, | |
| "loss": 0.7844, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 3.125e-05, | |
| "loss": 0.8477, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "eval_loss": 0.8244509696960449, | |
| "eval_runtime": 20.6023, | |
| "eval_samples_per_second": 3.106, | |
| "eval_steps_per_second": 0.388, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 3.38e-05, | |
| "loss": 0.786, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 3.6450000000000005e-05, | |
| "loss": 0.6724, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 3.9200000000000004e-05, | |
| "loss": 0.8566, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 4.205e-05, | |
| "loss": 0.6692, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 4.4999999999999996e-05, | |
| "loss": 0.7207, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "eval_loss": 0.6993950009346008, | |
| "eval_runtime": 20.6231, | |
| "eval_samples_per_second": 3.103, | |
| "eval_steps_per_second": 0.388, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 4.805e-05, | |
| "loss": 0.714, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 5.1200000000000004e-05, | |
| "loss": 0.6141, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 5.445000000000001e-05, | |
| "loss": 0.7061, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 5.780000000000001e-05, | |
| "loss": 0.652, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 6.125e-05, | |
| "loss": 0.4656, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "eval_loss": 0.5878281593322754, | |
| "eval_runtime": 20.6201, | |
| "eval_samples_per_second": 3.104, | |
| "eval_steps_per_second": 0.388, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 6.48e-05, | |
| "loss": 0.6225, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 6.845e-05, | |
| "loss": 0.4638, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 7.22e-05, | |
| "loss": 0.595, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 7.605e-05, | |
| "loss": 0.5618, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 8.000000000000002e-05, | |
| "loss": 0.4949, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "eval_loss": 0.4970114827156067, | |
| "eval_runtime": 20.6228, | |
| "eval_samples_per_second": 3.103, | |
| "eval_steps_per_second": 0.388, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 8.404999999999998e-05, | |
| "loss": 0.5576, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 8.819999999999999e-05, | |
| "loss": 0.5537, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 9.245e-05, | |
| "loss": 0.3734, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 9.68e-05, | |
| "loss": 0.4103, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 0.00010125000000000001, | |
| "loss": 0.3497, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "eval_loss": 0.42206770181655884, | |
| "eval_runtime": 20.6133, | |
| "eval_samples_per_second": 3.105, | |
| "eval_steps_per_second": 0.388, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 0.00010580000000000001, | |
| "loss": 0.4948, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 0.00011045, | |
| "loss": 0.2884, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 0.0001152, | |
| "loss": 0.4234, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 0.00012004999999999999, | |
| "loss": 0.3352, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 0.000125, | |
| "loss": 0.3288, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "eval_loss": 0.3672243356704712, | |
| "eval_runtime": 20.6204, | |
| "eval_samples_per_second": 3.104, | |
| "eval_steps_per_second": 0.388, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "learning_rate": 0.00013005, | |
| "loss": 0.3481, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "learning_rate": 0.0001352, | |
| "loss": 0.362, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 0.00014045000000000003, | |
| "loss": 0.344, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 0.00014580000000000002, | |
| "loss": 0.426, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 0.00015125000000000002, | |
| "loss": 0.3011, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "eval_loss": 0.32503682374954224, | |
| "eval_runtime": 20.6368, | |
| "eval_samples_per_second": 3.101, | |
| "eval_steps_per_second": 0.388, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "learning_rate": 0.00015680000000000002, | |
| "loss": 0.4011, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 0.00016245, | |
| "loss": 0.3796, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 0.0001682, | |
| "loss": 0.2726, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "learning_rate": 0.00017404999999999998, | |
| "loss": 0.3218, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "learning_rate": 0.00017999999999999998, | |
| "loss": 0.2648, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "eval_loss": 0.29000362753868103, | |
| "eval_runtime": 20.595, | |
| "eval_samples_per_second": 3.108, | |
| "eval_steps_per_second": 0.388, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "learning_rate": 0.00018605, | |
| "loss": 0.331, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "learning_rate": 0.0001922, | |
| "loss": 0.2442, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 3.32, | |
| "learning_rate": 0.00019845000000000003, | |
| "loss": 0.2839, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "learning_rate": 0.00020480000000000002, | |
| "loss": 0.3206, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "learning_rate": 0.00021125, | |
| "loss": 0.3084, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "eval_loss": 0.25911369919776917, | |
| "eval_runtime": 20.6071, | |
| "eval_samples_per_second": 3.106, | |
| "eval_steps_per_second": 0.388, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 3.47, | |
| "learning_rate": 0.00021780000000000004, | |
| "loss": 0.2479, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "learning_rate": 0.00022445000000000003, | |
| "loss": 0.3003, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 3.58, | |
| "learning_rate": 0.00023120000000000004, | |
| "loss": 0.2757, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 3.63, | |
| "learning_rate": 0.00023804999999999996, | |
| "loss": 0.1921, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 0.000245, | |
| "loss": 0.2696, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "eval_loss": 0.24587483704090118, | |
| "eval_runtime": 20.6172, | |
| "eval_samples_per_second": 3.104, | |
| "eval_steps_per_second": 0.388, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "learning_rate": 0.00025205000000000003, | |
| "loss": 0.2596, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "learning_rate": 0.0002592, | |
| "loss": 0.2386, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "learning_rate": 0.00026644999999999994, | |
| "loss": 0.2281, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "learning_rate": 0.0002738, | |
| "loss": 0.2538, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "learning_rate": 0.00028125000000000003, | |
| "loss": 0.2197, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "eval_loss": 0.22858355939388275, | |
| "eval_runtime": 20.6122, | |
| "eval_samples_per_second": 3.105, | |
| "eval_steps_per_second": 0.388, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 0.0002888, | |
| "loss": 0.182, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 0.00029645, | |
| "loss": 0.1985, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "learning_rate": 0.0003042, | |
| "loss": 0.1978, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "learning_rate": 0.0003120500000000001, | |
| "loss": 0.2283, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 4.21, | |
| "learning_rate": 0.0003200000000000001, | |
| "loss": 0.1905, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.21, | |
| "eval_loss": 0.21105319261550903, | |
| "eval_runtime": 20.595, | |
| "eval_samples_per_second": 3.108, | |
| "eval_steps_per_second": 0.388, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.26, | |
| "learning_rate": 0.0003280500000000001, | |
| "loss": 0.2222, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "learning_rate": 0.00033619999999999993, | |
| "loss": 0.2051, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 4.37, | |
| "learning_rate": 0.00034445, | |
| "loss": 0.2013, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 4.42, | |
| "learning_rate": 0.00035279999999999996, | |
| "loss": 0.1751, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 4.47, | |
| "learning_rate": 0.00036124999999999997, | |
| "loss": 0.1815, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 4.47, | |
| "eval_loss": 0.2084398865699768, | |
| "eval_runtime": 20.6383, | |
| "eval_samples_per_second": 3.101, | |
| "eval_steps_per_second": 0.388, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 4.53, | |
| "learning_rate": 0.0003698, | |
| "loss": 0.1252, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 4.58, | |
| "learning_rate": 0.00037845, | |
| "loss": 0.2354, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 4.63, | |
| "learning_rate": 0.0003872, | |
| "loss": 0.1651, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 4.68, | |
| "learning_rate": 0.00039605, | |
| "loss": 0.2001, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "learning_rate": 0.00040500000000000003, | |
| "loss": 0.2164, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "eval_loss": 0.21278557181358337, | |
| "eval_runtime": 20.6294, | |
| "eval_samples_per_second": 3.102, | |
| "eval_steps_per_second": 0.388, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 4.79, | |
| "learning_rate": 0.00041405000000000006, | |
| "loss": 0.155, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 4.84, | |
| "learning_rate": 0.00042320000000000004, | |
| "loss": 0.2454, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "learning_rate": 0.0004324500000000001, | |
| "loss": 0.1804, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "learning_rate": 0.0004418, | |
| "loss": 0.2498, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 0.00045125, | |
| "loss": 0.1412, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.2015175074338913, | |
| "eval_runtime": 20.5945, | |
| "eval_samples_per_second": 3.108, | |
| "eval_steps_per_second": 0.388, | |
| "step": 95 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 95, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 19, | |
| "total_flos": 9751829478899712.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |