| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.820359281437126, | |
| "eval_steps": 500, | |
| "global_step": 410, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.23952095808383234, | |
| "grad_norm": 2.16796875, | |
| "learning_rate": 0.00019970658011837404, | |
| "loss": 0.5442, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.47904191616766467, | |
| "grad_norm": 2.02734375, | |
| "learning_rate": 0.00019882804237803488, | |
| "loss": 0.5445, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.718562874251497, | |
| "grad_norm": 1.2587890625, | |
| "learning_rate": 0.00019736954238777792, | |
| "loss": 0.5499, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.9580838323353293, | |
| "grad_norm": 1.3486328125, | |
| "learning_rate": 0.00019533963920549306, | |
| "loss": 0.5429, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.1976047904191618, | |
| "grad_norm": 2.08984375, | |
| "learning_rate": 0.0001927502451102095, | |
| "loss": 0.5336, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.437125748502994, | |
| "grad_norm": 1.7734375, | |
| "learning_rate": 0.00018961655569610557, | |
| "loss": 0.5333, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.6766467065868262, | |
| "grad_norm": 2.63671875, | |
| "learning_rate": 0.00018595696069872013, | |
| "loss": 0.5401, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.9161676646706587, | |
| "grad_norm": 2.19921875, | |
| "learning_rate": 0.00018179293607667178, | |
| "loss": 0.5414, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.155688622754491, | |
| "grad_norm": 1.908203125, | |
| "learning_rate": 0.0001771489179821943, | |
| "loss": 0.5713, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.3952095808383236, | |
| "grad_norm": 4.09765625, | |
| "learning_rate": 0.0001720521593600787, | |
| "loss": 0.538, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.6347305389221556, | |
| "grad_norm": 2.0390625, | |
| "learning_rate": 0.00016653257001655652, | |
| "loss": 0.5199, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.874251497005988, | |
| "grad_norm": 1.8154296875, | |
| "learning_rate": 0.0001606225410966638, | |
| "loss": 0.5204, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.1137724550898205, | |
| "grad_norm": 1.9189453125, | |
| "learning_rate": 0.00015435675500012212, | |
| "loss": 0.5396, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.3532934131736525, | |
| "grad_norm": 1.8896484375, | |
| "learning_rate": 0.0001477719818512263, | |
| "loss": 0.5294, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.592814371257485, | |
| "grad_norm": 2.302734375, | |
| "learning_rate": 0.00014090686371713402, | |
| "loss": 0.5394, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.8323353293413174, | |
| "grad_norm": 2.16796875, | |
| "learning_rate": 0.00013380168784085027, | |
| "loss": 0.5279, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 4.07185628742515, | |
| "grad_norm": 2.11328125, | |
| "learning_rate": 0.0001264981502196662, | |
| "loss": 0.5182, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 4.311377245508982, | |
| "grad_norm": 1.9189453125, | |
| "learning_rate": 0.00011903911091646684, | |
| "loss": 0.5304, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 4.550898203592815, | |
| "grad_norm": 2.48046875, | |
| "learning_rate": 0.00011146834253984006, | |
| "loss": 0.5386, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 4.790419161676647, | |
| "grad_norm": 2.62890625, | |
| "learning_rate": 0.00010383027336900355, | |
| "loss": 0.5276, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 5.029940119760479, | |
| "grad_norm": 3.021484375, | |
| "learning_rate": 9.616972663099647e-05, | |
| "loss": 0.5229, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 5.269461077844311, | |
| "grad_norm": 3.259765625, | |
| "learning_rate": 8.853165746015997e-05, | |
| "loss": 0.5243, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 5.508982035928144, | |
| "grad_norm": 2.876953125, | |
| "learning_rate": 8.096088908353315e-05, | |
| "loss": 0.5206, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 5.748502994011976, | |
| "grad_norm": 3.349609375, | |
| "learning_rate": 7.350184978033386e-05, | |
| "loss": 0.5374, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 5.9880239520958085, | |
| "grad_norm": 3.791015625, | |
| "learning_rate": 6.619831215914974e-05, | |
| "loss": 0.5258, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 6.227544910179641, | |
| "grad_norm": 3.240234375, | |
| "learning_rate": 5.909313628286601e-05, | |
| "loss": 0.5166, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 6.467065868263473, | |
| "grad_norm": 3.248046875, | |
| "learning_rate": 5.222801814877369e-05, | |
| "loss": 0.5329, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 6.706586826347305, | |
| "grad_norm": 2.83984375, | |
| "learning_rate": 4.56432449998779e-05, | |
| "loss": 0.5117, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 6.946107784431137, | |
| "grad_norm": 4.03515625, | |
| "learning_rate": 3.937745890333623e-05, | |
| "loss": 0.5269, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 7.18562874251497, | |
| "grad_norm": 4.08203125, | |
| "learning_rate": 3.346742998344348e-05, | |
| "loss": 0.5299, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 7.425149700598802, | |
| "grad_norm": 3.03125, | |
| "learning_rate": 2.794784063992131e-05, | |
| "loss": 0.5217, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 7.664670658682635, | |
| "grad_norm": 3.00390625, | |
| "learning_rate": 2.2851082017805703e-05, | |
| "loss": 0.5241, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 7.904191616766467, | |
| "grad_norm": 3.27734375, | |
| "learning_rate": 1.8207063923328237e-05, | |
| "loss": 0.5066, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 8.1437125748503, | |
| "grad_norm": 2.927734375, | |
| "learning_rate": 1.4043039301279903e-05, | |
| "loss": 0.5111, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 8.383233532934131, | |
| "grad_norm": 3.3984375, | |
| "learning_rate": 1.0383444303894452e-05, | |
| "loss": 0.5137, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 8.622754491017965, | |
| "grad_norm": 3.4375, | |
| "learning_rate": 7.249754889790539e-06, | |
| "loss": 0.5279, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 8.862275449101796, | |
| "grad_norm": 3.19140625, | |
| "learning_rate": 4.660360794506946e-06, | |
| "loss": 0.5125, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 9.10179640718563, | |
| "grad_norm": 4.23046875, | |
| "learning_rate": 2.6304576122221035e-06, | |
| "loss": 0.5141, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 9.341317365269461, | |
| "grad_norm": 2.779296875, | |
| "learning_rate": 1.1719576219651585e-06, | |
| "loss": 0.5172, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 9.580838323353294, | |
| "grad_norm": 3.619140625, | |
| "learning_rate": 2.934198816259559e-07, | |
| "loss": 0.5389, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 9.820359281437126, | |
| "grad_norm": 4.30078125, | |
| "learning_rate": 0.0, | |
| "loss": 0.5024, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 9.820359281437126, | |
| "step": 410, | |
| "total_flos": 2.67670788243456e+16, | |
| "train_loss": 0.5285330202521348, | |
| "train_runtime": 355.5253, | |
| "train_samples_per_second": 4.697, | |
| "train_steps_per_second": 1.153 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 410, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 2.67670788243456e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |