| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.9411764705882355, | |
| "eval_steps": 20, | |
| "global_step": 111, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05378151260504202, | |
| "grad_norm": 0.012224463745951653, | |
| "learning_rate": 0.001981818181818182, | |
| "loss": 2.4733, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.10756302521008404, | |
| "grad_norm": 0.04710804298520088, | |
| "learning_rate": 0.0019454545454545456, | |
| "loss": 2.4443, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.16134453781512606, | |
| "grad_norm": 0.09238269925117493, | |
| "learning_rate": 0.0019090909090909091, | |
| "loss": 2.3956, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.21512605042016808, | |
| "grad_norm": 0.17334744334220886, | |
| "learning_rate": 0.0018727272727272729, | |
| "loss": 2.4043, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.2689075630252101, | |
| "grad_norm": 0.11766365170478821, | |
| "learning_rate": 0.0018363636363636364, | |
| "loss": 2.385, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.3226890756302521, | |
| "grad_norm": 0.1385774314403534, | |
| "learning_rate": 0.0018000000000000002, | |
| "loss": 2.3933, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.3764705882352941, | |
| "grad_norm": 0.1210022047162056, | |
| "learning_rate": 0.0017636363636363637, | |
| "loss": 2.3733, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.43025210084033616, | |
| "grad_norm": 0.11510306596755981, | |
| "learning_rate": 0.0017272727272727272, | |
| "loss": 2.3297, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.48403361344537815, | |
| "grad_norm": 0.08369912207126617, | |
| "learning_rate": 0.001690909090909091, | |
| "loss": 2.351, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.5378151260504201, | |
| "grad_norm": 0.09298688918352127, | |
| "learning_rate": 0.0016545454545454545, | |
| "loss": 2.3175, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.5378151260504201, | |
| "eval_loss": 2.3251912593841553, | |
| "eval_runtime": 84.2914, | |
| "eval_samples_per_second": 14.118, | |
| "eval_steps_per_second": 1.768, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.5915966386554622, | |
| "grad_norm": 0.10441266000270844, | |
| "learning_rate": 0.0016181818181818183, | |
| "loss": 2.3643, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.6453781512605042, | |
| "grad_norm": 0.09343012422323227, | |
| "learning_rate": 0.0015818181818181818, | |
| "loss": 2.3391, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.6991596638655462, | |
| "grad_norm": 0.09008985757827759, | |
| "learning_rate": 0.0015454545454545454, | |
| "loss": 2.2984, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.7529411764705882, | |
| "grad_norm": 0.08069847524166107, | |
| "learning_rate": 0.0015090909090909091, | |
| "loss": 2.3202, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.8067226890756303, | |
| "grad_norm": 0.08655106276273727, | |
| "learning_rate": 0.0014727272727272727, | |
| "loss": 2.3438, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.8605042016806723, | |
| "grad_norm": 0.08203998953104019, | |
| "learning_rate": 0.0014363636363636362, | |
| "loss": 2.2862, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.9142857142857143, | |
| "grad_norm": 0.10055471211671829, | |
| "learning_rate": 0.0014, | |
| "loss": 2.3448, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.9680672268907563, | |
| "grad_norm": 0.08500978350639343, | |
| "learning_rate": 0.0013636363636363635, | |
| "loss": 2.3221, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.16502036154270172, | |
| "learning_rate": 0.0013272727272727275, | |
| "loss": 2.3438, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.053781512605042, | |
| "grad_norm": 0.08134379237890244, | |
| "learning_rate": 0.001290909090909091, | |
| "loss": 2.3075, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.053781512605042, | |
| "eval_loss": 2.299827814102173, | |
| "eval_runtime": 84.1245, | |
| "eval_samples_per_second": 14.146, | |
| "eval_steps_per_second": 1.771, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.107563025210084, | |
| "grad_norm": 0.09189953655004501, | |
| "learning_rate": 0.0012545454545454546, | |
| "loss": 2.2457, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.1613445378151261, | |
| "grad_norm": 0.09041959792375565, | |
| "learning_rate": 0.0012181818181818183, | |
| "loss": 2.2977, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.2151260504201682, | |
| "grad_norm": 0.08456366509199142, | |
| "learning_rate": 0.0011818181818181819, | |
| "loss": 2.2843, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.26890756302521, | |
| "grad_norm": 0.08097781240940094, | |
| "learning_rate": 0.0011454545454545454, | |
| "loss": 2.2328, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.322689075630252, | |
| "grad_norm": 0.10243827849626541, | |
| "learning_rate": 0.0011090909090909092, | |
| "loss": 2.254, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.3764705882352941, | |
| "grad_norm": 0.09242815524339676, | |
| "learning_rate": 0.0010727272727272727, | |
| "loss": 2.3295, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.4302521008403362, | |
| "grad_norm": 0.09403648227453232, | |
| "learning_rate": 0.0010363636363636365, | |
| "loss": 2.2749, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.4840336134453782, | |
| "grad_norm": 0.09187959879636765, | |
| "learning_rate": 0.001, | |
| "loss": 2.2606, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.53781512605042, | |
| "grad_norm": 0.09116198122501373, | |
| "learning_rate": 0.0009636363636363637, | |
| "loss": 2.2676, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 1.5915966386554623, | |
| "grad_norm": 0.08270075172185898, | |
| "learning_rate": 0.0009272727272727273, | |
| "loss": 2.2741, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.5915966386554623, | |
| "eval_loss": 2.3082737922668457, | |
| "eval_runtime": 84.2709, | |
| "eval_samples_per_second": 14.121, | |
| "eval_steps_per_second": 1.768, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.6453781512605041, | |
| "grad_norm": 0.09275200217962265, | |
| "learning_rate": 0.0008909090909090909, | |
| "loss": 2.2582, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 1.6991596638655462, | |
| "grad_norm": 0.09241969138383865, | |
| "learning_rate": 0.0008545454545454545, | |
| "loss": 2.2554, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 1.7529411764705882, | |
| "grad_norm": 0.08338718116283417, | |
| "learning_rate": 0.0008181818181818183, | |
| "loss": 2.2244, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.8067226890756303, | |
| "grad_norm": 0.09568168222904205, | |
| "learning_rate": 0.0007818181818181819, | |
| "loss": 2.2719, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 1.8605042016806723, | |
| "grad_norm": 0.0905410498380661, | |
| "learning_rate": 0.0007454545454545455, | |
| "loss": 2.2505, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.9142857142857141, | |
| "grad_norm": 0.08841802924871445, | |
| "learning_rate": 0.0007090909090909091, | |
| "loss": 2.3005, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.9680672268907564, | |
| "grad_norm": 0.09013470262289047, | |
| "learning_rate": 0.0006727272727272728, | |
| "loss": 2.2682, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.19737772643566132, | |
| "learning_rate": 0.0006363636363636364, | |
| "loss": 2.3476, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 2.053781512605042, | |
| "grad_norm": 0.0839110016822815, | |
| "learning_rate": 0.0006, | |
| "loss": 2.2338, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 2.107563025210084, | |
| "grad_norm": 0.10582801699638367, | |
| "learning_rate": 0.0005636363636363636, | |
| "loss": 2.2257, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.107563025210084, | |
| "eval_loss": 2.304074764251709, | |
| "eval_runtime": 84.3347, | |
| "eval_samples_per_second": 14.11, | |
| "eval_steps_per_second": 1.767, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.161344537815126, | |
| "grad_norm": 0.09145358949899673, | |
| "learning_rate": 0.0005272727272727272, | |
| "loss": 2.2488, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 2.215126050420168, | |
| "grad_norm": 0.08459240943193436, | |
| "learning_rate": 0.0004909090909090909, | |
| "loss": 2.2518, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 2.26890756302521, | |
| "grad_norm": 0.09590018540620804, | |
| "learning_rate": 0.00045454545454545455, | |
| "loss": 2.2324, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 2.3226890756302523, | |
| "grad_norm": 0.10032965242862701, | |
| "learning_rate": 0.00041818181818181814, | |
| "loss": 2.2099, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 2.376470588235294, | |
| "grad_norm": 0.09092257171869278, | |
| "learning_rate": 0.00038181818181818184, | |
| "loss": 2.2077, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.4302521008403364, | |
| "grad_norm": 0.10066290944814682, | |
| "learning_rate": 0.00034545454545454544, | |
| "loss": 2.2629, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 2.484033613445378, | |
| "grad_norm": 0.0973694771528244, | |
| "learning_rate": 0.0003090909090909091, | |
| "loss": 2.2292, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 2.53781512605042, | |
| "grad_norm": 0.09254106879234314, | |
| "learning_rate": 0.00027272727272727274, | |
| "loss": 2.1923, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 2.5915966386554623, | |
| "grad_norm": 0.10056042671203613, | |
| "learning_rate": 0.00023636363636363636, | |
| "loss": 2.2445, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 2.645378151260504, | |
| "grad_norm": 0.09601625055074692, | |
| "learning_rate": 0.0002, | |
| "loss": 2.2605, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.645378151260504, | |
| "eval_loss": 2.3115394115448, | |
| "eval_runtime": 84.2807, | |
| "eval_samples_per_second": 14.119, | |
| "eval_steps_per_second": 1.768, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.6991596638655464, | |
| "grad_norm": 0.09498832374811172, | |
| "learning_rate": 0.00016363636363636363, | |
| "loss": 2.215, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 2.7529411764705882, | |
| "grad_norm": 0.09191343188285828, | |
| "learning_rate": 0.00012727272727272725, | |
| "loss": 2.2116, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 2.80672268907563, | |
| "grad_norm": 0.10717286169528961, | |
| "learning_rate": 9.090909090909092e-05, | |
| "loss": 2.2435, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 2.8605042016806723, | |
| "grad_norm": 0.09715902805328369, | |
| "learning_rate": 5.4545454545454546e-05, | |
| "loss": 2.2196, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 2.914285714285714, | |
| "grad_norm": 0.10500436276197433, | |
| "learning_rate": 1.8181818181818182e-05, | |
| "loss": 2.2351, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.9411764705882355, | |
| "step": 111, | |
| "total_flos": 8.1776874848256e+17, | |
| "train_loss": 2.2894913076280474, | |
| "train_runtime": 2825.7191, | |
| "train_samples_per_second": 10.107, | |
| "train_steps_per_second": 0.039 | |
| }, | |
| { | |
| "epoch": 2.9411764705882355, | |
| "eval_loss": 2.311664342880249, | |
| "eval_runtime": 84.4111, | |
| "eval_samples_per_second": 14.098, | |
| "eval_steps_per_second": 1.765, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 2.9411764705882355, | |
| "eval_loss": 2.3323311805725098, | |
| "eval_runtime": 84.1513, | |
| "eval_samples_per_second": 14.141, | |
| "eval_steps_per_second": 1.771, | |
| "step": 111 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 111, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 8.1776874848256e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |