| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.7391304347826086, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.017391304347826087, | |
| "eval_loss": 1.59055495262146, | |
| "eval_runtime": 2.6995, | |
| "eval_samples_per_second": 35.932, | |
| "eval_steps_per_second": 4.816, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.05217391304347826, | |
| "grad_norm": 0.5041409134864807, | |
| "learning_rate": 1.5e-05, | |
| "loss": 1.7739, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.10434782608695652, | |
| "grad_norm": 0.5414333343505859, | |
| "learning_rate": 3e-05, | |
| "loss": 1.402, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.1565217391304348, | |
| "grad_norm": 0.3529357612133026, | |
| "learning_rate": 4.5e-05, | |
| "loss": 1.6939, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.1565217391304348, | |
| "eval_loss": 1.5777380466461182, | |
| "eval_runtime": 2.6751, | |
| "eval_samples_per_second": 36.26, | |
| "eval_steps_per_second": 4.86, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.20869565217391303, | |
| "grad_norm": 0.3947656452655792, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 1.7722, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.2608695652173913, | |
| "grad_norm": 0.4697956442832947, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 1.6984, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.3130434782608696, | |
| "grad_norm": 0.34927043318748474, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 1.7339, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.3130434782608696, | |
| "eval_loss": 1.4486348628997803, | |
| "eval_runtime": 2.683, | |
| "eval_samples_per_second": 36.153, | |
| "eval_steps_per_second": 4.845, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.3652173913043478, | |
| "grad_norm": 0.4831562042236328, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 1.7264, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.41739130434782606, | |
| "grad_norm": 0.56829833984375, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 1.7197, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.46956521739130436, | |
| "grad_norm": 0.5362353920936584, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 1.5556, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.46956521739130436, | |
| "eval_loss": 1.2610658407211304, | |
| "eval_runtime": 2.6707, | |
| "eval_samples_per_second": 36.32, | |
| "eval_steps_per_second": 4.868, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.5217391304347826, | |
| "grad_norm": 0.5250989198684692, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 1.5176, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.5739130434782609, | |
| "grad_norm": 0.4496021866798401, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 1.2318, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.6260869565217392, | |
| "grad_norm": 0.5903450846672058, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 1.5067, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.6260869565217392, | |
| "eval_loss": 1.09618079662323, | |
| "eval_runtime": 2.705, | |
| "eval_samples_per_second": 35.859, | |
| "eval_steps_per_second": 4.806, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.6782608695652174, | |
| "grad_norm": 0.5637620687484741, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 1.2329, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.7304347826086957, | |
| "grad_norm": 0.5909444689750671, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 1.1362, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.782608695652174, | |
| "grad_norm": 0.4471411406993866, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 1.3341, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.782608695652174, | |
| "eval_loss": 1.0226554870605469, | |
| "eval_runtime": 2.6701, | |
| "eval_samples_per_second": 36.328, | |
| "eval_steps_per_second": 4.869, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.8347826086956521, | |
| "grad_norm": 0.4761001169681549, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 1.0345, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.8869565217391304, | |
| "grad_norm": 0.4806910753250122, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 0.9969, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.9391304347826087, | |
| "grad_norm": 0.6003803014755249, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 1.2715, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.9391304347826087, | |
| "eval_loss": 0.9930722117424011, | |
| "eval_runtime": 2.6757, | |
| "eval_samples_per_second": 36.252, | |
| "eval_steps_per_second": 4.859, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.991304347826087, | |
| "grad_norm": 0.5368942618370056, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 1.1923, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 1.0434782608695652, | |
| "grad_norm": 0.5342758893966675, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 1.0875, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.0956521739130434, | |
| "grad_norm": 0.4160667359828949, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 1.3013, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.0956521739130434, | |
| "eval_loss": 0.9771125912666321, | |
| "eval_runtime": 2.6789, | |
| "eval_samples_per_second": 36.209, | |
| "eval_steps_per_second": 4.853, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.1478260869565218, | |
| "grad_norm": 0.38854989409446716, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 1.4154, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 0.5519515872001648, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 1.1699, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 1.2521739130434781, | |
| "grad_norm": 0.3825100362300873, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 1.188, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.2521739130434781, | |
| "eval_loss": 0.9695695638656616, | |
| "eval_runtime": 2.6795, | |
| "eval_samples_per_second": 36.201, | |
| "eval_steps_per_second": 4.852, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.3043478260869565, | |
| "grad_norm": 0.47509416937828064, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 1.2668, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.3565217391304347, | |
| "grad_norm": 0.36375659704208374, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 1.3312, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 1.4086956521739131, | |
| "grad_norm": 0.5056440830230713, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 1.2621, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.4086956521739131, | |
| "eval_loss": 0.9655399918556213, | |
| "eval_runtime": 2.671, | |
| "eval_samples_per_second": 36.316, | |
| "eval_steps_per_second": 4.867, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.4608695652173913, | |
| "grad_norm": 0.48886847496032715, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 0.968, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.5130434782608697, | |
| "grad_norm": 0.44092217087745667, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 1.467, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 1.5652173913043477, | |
| "grad_norm": 0.426158607006073, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 1.3827, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.5652173913043477, | |
| "eval_loss": 0.9642850756645203, | |
| "eval_runtime": 2.6781, | |
| "eval_samples_per_second": 36.22, | |
| "eval_steps_per_second": 4.854, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.617391304347826, | |
| "grad_norm": 0.4980301558971405, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 1.164, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 1.6695652173913045, | |
| "grad_norm": 0.4769861102104187, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 0.7116, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.7217391304347827, | |
| "grad_norm": 0.4263245463371277, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 1.272, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 1.7217391304347827, | |
| "eval_loss": 0.9631911516189575, | |
| "eval_runtime": 2.7065, | |
| "eval_samples_per_second": 35.84, | |
| "eval_steps_per_second": 4.803, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "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": 2.6378466323595264e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |