| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.3436426116838488, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.003436426116838488, | |
| "eval_loss": 2.314373254776001, | |
| "eval_runtime": 13.6737, | |
| "eval_samples_per_second": 35.835, | |
| "eval_steps_per_second": 4.534, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.010309278350515464, | |
| "grad_norm": 3.7942211627960205, | |
| "learning_rate": 1.5e-05, | |
| "loss": 9.2256, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.020618556701030927, | |
| "grad_norm": 3.6376962661743164, | |
| "learning_rate": 3e-05, | |
| "loss": 9.3204, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.030927835051546393, | |
| "grad_norm": 2.9720754623413086, | |
| "learning_rate": 4.5e-05, | |
| "loss": 8.9964, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.030927835051546393, | |
| "eval_loss": 2.239830732345581, | |
| "eval_runtime": 13.9126, | |
| "eval_samples_per_second": 35.22, | |
| "eval_steps_per_second": 4.456, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.041237113402061855, | |
| "grad_norm": 2.528602123260498, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 8.8083, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.05154639175257732, | |
| "grad_norm": 2.454768180847168, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 8.7321, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.061855670103092786, | |
| "grad_norm": 2.124577045440674, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 8.3309, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.061855670103092786, | |
| "eval_loss": 2.1040642261505127, | |
| "eval_runtime": 13.893, | |
| "eval_samples_per_second": 35.27, | |
| "eval_steps_per_second": 4.463, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.07216494845360824, | |
| "grad_norm": 2.270033597946167, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 8.1226, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.08247422680412371, | |
| "grad_norm": 2.5098090171813965, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 8.2941, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.09278350515463918, | |
| "grad_norm": 2.4826111793518066, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 8.0886, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.09278350515463918, | |
| "eval_loss": 2.0421814918518066, | |
| "eval_runtime": 13.8972, | |
| "eval_samples_per_second": 35.259, | |
| "eval_steps_per_second": 4.461, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.10309278350515463, | |
| "grad_norm": 2.2102906703948975, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 8.2557, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.1134020618556701, | |
| "grad_norm": 2.479996919631958, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 7.7965, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.12371134020618557, | |
| "grad_norm": 2.1732442378997803, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 7.8037, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.12371134020618557, | |
| "eval_loss": 2.005703926086426, | |
| "eval_runtime": 13.7059, | |
| "eval_samples_per_second": 35.751, | |
| "eval_steps_per_second": 4.524, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.13402061855670103, | |
| "grad_norm": 2.0529301166534424, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 7.7439, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.14432989690721648, | |
| "grad_norm": 2.04811954498291, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 8.0662, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.15463917525773196, | |
| "grad_norm": 2.075685977935791, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 7.8449, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.15463917525773196, | |
| "eval_loss": 1.9820556640625, | |
| "eval_runtime": 13.9153, | |
| "eval_samples_per_second": 35.213, | |
| "eval_steps_per_second": 4.456, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.16494845360824742, | |
| "grad_norm": 2.139718532562256, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 7.6805, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.17525773195876287, | |
| "grad_norm": 2.1034138202667236, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 7.873, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.18556701030927836, | |
| "grad_norm": 2.083611011505127, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 7.9978, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.18556701030927836, | |
| "eval_loss": 1.9646066427230835, | |
| "eval_runtime": 13.9213, | |
| "eval_samples_per_second": 35.198, | |
| "eval_steps_per_second": 4.454, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.1958762886597938, | |
| "grad_norm": 2.087646245956421, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 7.7954, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.20618556701030927, | |
| "grad_norm": 2.0635478496551514, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 7.8197, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.21649484536082475, | |
| "grad_norm": 2.268955707550049, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 7.5581, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.21649484536082475, | |
| "eval_loss": 1.9571253061294556, | |
| "eval_runtime": 13.9191, | |
| "eval_samples_per_second": 35.203, | |
| "eval_steps_per_second": 4.454, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.2268041237113402, | |
| "grad_norm": 1.9224090576171875, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 7.884, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.23711340206185566, | |
| "grad_norm": 2.2408390045166016, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 7.8544, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.24742268041237114, | |
| "grad_norm": 2.2292492389678955, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 7.7959, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.24742268041237114, | |
| "eval_loss": 1.95169997215271, | |
| "eval_runtime": 13.9247, | |
| "eval_samples_per_second": 35.189, | |
| "eval_steps_per_second": 4.453, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.25773195876288657, | |
| "grad_norm": 2.050633430480957, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 7.8615, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.26804123711340205, | |
| "grad_norm": 2.713322639465332, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 7.762, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.27835051546391754, | |
| "grad_norm": 1.9455171823501587, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 7.4536, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.27835051546391754, | |
| "eval_loss": 1.9475512504577637, | |
| "eval_runtime": 13.9146, | |
| "eval_samples_per_second": 35.215, | |
| "eval_steps_per_second": 4.456, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.28865979381443296, | |
| "grad_norm": 2.1871798038482666, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 7.9459, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.29896907216494845, | |
| "grad_norm": 2.183077812194824, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 7.5762, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.30927835051546393, | |
| "grad_norm": 1.9717400074005127, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 7.7221, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.30927835051546393, | |
| "eval_loss": 1.9463396072387695, | |
| "eval_runtime": 13.9155, | |
| "eval_samples_per_second": 35.213, | |
| "eval_steps_per_second": 4.455, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.31958762886597936, | |
| "grad_norm": 2.0875370502471924, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 7.8852, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.32989690721649484, | |
| "grad_norm": 2.3906922340393066, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 7.884, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.3402061855670103, | |
| "grad_norm": 1.9967540502548218, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 7.6559, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.3402061855670103, | |
| "eval_loss": 1.9455057382583618, | |
| "eval_runtime": 13.7004, | |
| "eval_samples_per_second": 35.765, | |
| "eval_steps_per_second": 4.525, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6567981899317248.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |