| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.004524323896347739, | |
| "eval_steps": 50, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 2.26216194817387e-05, | |
| "eval_loss": 4.674833297729492, | |
| "eval_runtime": 1630.3176, | |
| "eval_samples_per_second": 11.417, | |
| "eval_steps_per_second": 5.709, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00011310809740869349, | |
| "grad_norm": 1.904850959777832, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0451, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00022621619481738697, | |
| "grad_norm": 3.487840414047241, | |
| "learning_rate": 0.0001, | |
| "loss": 1.0945, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.00033932429222608047, | |
| "grad_norm": 7.243443965911865, | |
| "learning_rate": 9.98292246503335e-05, | |
| "loss": 1.5305, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.00045243238963477394, | |
| "grad_norm": 9.466878890991211, | |
| "learning_rate": 9.931806517013612e-05, | |
| "loss": 0.8418, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0005655404870434674, | |
| "grad_norm": 14.048686981201172, | |
| "learning_rate": 9.847001329696653e-05, | |
| "loss": 0.612, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0006786485844521609, | |
| "grad_norm": 12.626312255859375, | |
| "learning_rate": 9.729086208503174e-05, | |
| "loss": 0.5316, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0007917566818608545, | |
| "grad_norm": 1.7504353523254395, | |
| "learning_rate": 9.578866633275288e-05, | |
| "loss": 0.2136, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.0009048647792695479, | |
| "grad_norm": 20.777067184448242, | |
| "learning_rate": 9.397368756032445e-05, | |
| "loss": 0.4962, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.0010179728766782413, | |
| "grad_norm": 15.338814735412598, | |
| "learning_rate": 9.185832391312644e-05, | |
| "loss": 0.4096, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0011310809740869348, | |
| "grad_norm": 16.687267303466797, | |
| "learning_rate": 8.945702546981969e-05, | |
| "loss": 0.9656, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.0011310809740869348, | |
| "eval_loss": 0.6156600713729858, | |
| "eval_runtime": 1638.7944, | |
| "eval_samples_per_second": 11.358, | |
| "eval_steps_per_second": 5.679, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.0012441890714956284, | |
| "grad_norm": 2.586226224899292, | |
| "learning_rate": 8.678619553365659e-05, | |
| "loss": 0.7048, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.0013572971689043219, | |
| "grad_norm": 5.269473552703857, | |
| "learning_rate": 8.386407858128706e-05, | |
| "loss": 0.7112, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.0014704052663130154, | |
| "grad_norm": 12.071560859680176, | |
| "learning_rate": 8.07106356344834e-05, | |
| "loss": 0.8531, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.001583513363721709, | |
| "grad_norm": 7.238372325897217, | |
| "learning_rate": 7.734740790612136e-05, | |
| "loss": 0.6405, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.0016966214611304022, | |
| "grad_norm": 7.999783039093018, | |
| "learning_rate": 7.379736965185368e-05, | |
| "loss": 0.5897, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.0018097295585390958, | |
| "grad_norm": 4.573731422424316, | |
| "learning_rate": 7.008477123264848e-05, | |
| "loss": 0.4493, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.0019228376559477893, | |
| "grad_norm": 9.74732780456543, | |
| "learning_rate": 6.623497346023418e-05, | |
| "loss": 0.3617, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.0020359457533564826, | |
| "grad_norm": 10.304098129272461, | |
| "learning_rate": 6.227427435703997e-05, | |
| "loss": 0.3788, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.0021490538507651764, | |
| "grad_norm": 4.553127765655518, | |
| "learning_rate": 5.8229729514036705e-05, | |
| "loss": 0.2911, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.0022621619481738697, | |
| "grad_norm": 12.018383979797363, | |
| "learning_rate": 5.4128967273616625e-05, | |
| "loss": 0.9804, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.0022621619481738697, | |
| "eval_loss": 0.46400874853134155, | |
| "eval_runtime": 1639.1544, | |
| "eval_samples_per_second": 11.355, | |
| "eval_steps_per_second": 5.678, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.0023752700455825634, | |
| "grad_norm": 2.609111785888672, | |
| "learning_rate": 5e-05, | |
| "loss": 0.5825, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.0024883781429912567, | |
| "grad_norm": 1.579280138015747, | |
| "learning_rate": 4.5871032726383386e-05, | |
| "loss": 0.542, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.00260148624039995, | |
| "grad_norm": 9.289449691772461, | |
| "learning_rate": 4.17702704859633e-05, | |
| "loss": 0.8798, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.0027145943378086438, | |
| "grad_norm": 5.802079677581787, | |
| "learning_rate": 3.772572564296005e-05, | |
| "loss": 0.7821, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.002827702435217337, | |
| "grad_norm": 5.292053699493408, | |
| "learning_rate": 3.3765026539765834e-05, | |
| "loss": 0.5794, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.002940810532626031, | |
| "grad_norm": 10.068510055541992, | |
| "learning_rate": 2.991522876735154e-05, | |
| "loss": 0.2694, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.003053918630034724, | |
| "grad_norm": 12.650135040283203, | |
| "learning_rate": 2.6202630348146324e-05, | |
| "loss": 0.2636, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.003167026727443418, | |
| "grad_norm": 0.9717761278152466, | |
| "learning_rate": 2.2652592093878666e-05, | |
| "loss": 0.2023, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.003280134824852111, | |
| "grad_norm": 5.5953288078308105, | |
| "learning_rate": 1.928936436551661e-05, | |
| "loss": 0.4384, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.0033932429222608045, | |
| "grad_norm": 7.228976726531982, | |
| "learning_rate": 1.6135921418712956e-05, | |
| "loss": 0.6151, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.0033932429222608045, | |
| "eval_loss": 0.42118245363235474, | |
| "eval_runtime": 1638.6941, | |
| "eval_samples_per_second": 11.358, | |
| "eval_steps_per_second": 5.68, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.0035063510196694982, | |
| "grad_norm": 3.614445209503174, | |
| "learning_rate": 1.3213804466343421e-05, | |
| "loss": 0.5205, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.0036194591170781915, | |
| "grad_norm": 2.2330234050750732, | |
| "learning_rate": 1.0542974530180327e-05, | |
| "loss": 0.539, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.0037325672144868853, | |
| "grad_norm": 7.5552802085876465, | |
| "learning_rate": 8.141676086873572e-06, | |
| "loss": 0.6344, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.0038456753118955786, | |
| "grad_norm": 4.733217716217041, | |
| "learning_rate": 6.026312439675552e-06, | |
| "loss": 0.4547, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.003958783409304272, | |
| "grad_norm": 2.8408470153808594, | |
| "learning_rate": 4.2113336672471245e-06, | |
| "loss": 0.2753, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.004071891506712965, | |
| "grad_norm": 13.729647636413574, | |
| "learning_rate": 2.7091379149682685e-06, | |
| "loss": 0.2504, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.004184999604121659, | |
| "grad_norm": 2.171924352645874, | |
| "learning_rate": 1.5299867030334814e-06, | |
| "loss": 0.1222, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.004298107701530353, | |
| "grad_norm": 0.47484132647514343, | |
| "learning_rate": 6.819348298638839e-07, | |
| "loss": 0.1661, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.0044112157989390464, | |
| "grad_norm": 5.313414573669434, | |
| "learning_rate": 1.7077534966650766e-07, | |
| "loss": 0.4323, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.004524323896347739, | |
| "grad_norm": 5.613385200500488, | |
| "learning_rate": 0.0, | |
| "loss": 0.3597, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.004524323896347739, | |
| "eval_loss": 0.41343122720718384, | |
| "eval_runtime": 1638.9598, | |
| "eval_samples_per_second": 11.357, | |
| "eval_steps_per_second": 5.679, | |
| "step": 200 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "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": 7.209325172883456e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |