| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 680, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.14705882352941177, | |
| "grad_norm": 5.515628814697266, | |
| "learning_rate": 2.9411764705882354e-05, | |
| "loss": 1.1436, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.29411764705882354, | |
| "grad_norm": 2.5779266357421875, | |
| "learning_rate": 5.882352941176471e-05, | |
| "loss": 0.4215, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.4411764705882353, | |
| "grad_norm": 1.5243560075759888, | |
| "learning_rate": 8.823529411764706e-05, | |
| "loss": 0.2531, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.5882352941176471, | |
| "grad_norm": 1.0052002668380737, | |
| "learning_rate": 9.997871633546257e-05, | |
| "loss": 0.1957, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.7352941176470589, | |
| "grad_norm": 1.7932289838790894, | |
| "learning_rate": 9.98487151097676e-05, | |
| "loss": 0.1667, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.8823529411764706, | |
| "grad_norm": 0.6952548623085022, | |
| "learning_rate": 9.960084393841355e-05, | |
| "loss": 0.1425, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.0294117647058822, | |
| "grad_norm": 0.944196343421936, | |
| "learning_rate": 9.923568892600578e-05, | |
| "loss": 0.1222, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.1764705882352942, | |
| "grad_norm": 0.6027107238769531, | |
| "learning_rate": 9.875411350104744e-05, | |
| "loss": 0.1081, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.3235294117647058, | |
| "grad_norm": 0.7133479714393616, | |
| "learning_rate": 9.815725637431662e-05, | |
| "loss": 0.1004, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.4705882352941178, | |
| "grad_norm": 0.6044264435768127, | |
| "learning_rate": 9.744652884632406e-05, | |
| "loss": 0.0935, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.6176470588235294, | |
| "grad_norm": 0.5952054858207703, | |
| "learning_rate": 9.662361147021779e-05, | |
| "loss": 0.0938, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.7647058823529411, | |
| "grad_norm": 0.7661593556404114, | |
| "learning_rate": 9.569045007802559e-05, | |
| "loss": 0.0876, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.9117647058823528, | |
| "grad_norm": 0.5203129649162292, | |
| "learning_rate": 9.464925117963133e-05, | |
| "loss": 0.0799, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.0588235294117645, | |
| "grad_norm": 1.392182469367981, | |
| "learning_rate": 9.35024767453647e-05, | |
| "loss": 0.0848, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.2058823529411766, | |
| "grad_norm": 0.44986793398857117, | |
| "learning_rate": 9.225283838454111e-05, | |
| "loss": 0.0789, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.3529411764705883, | |
| "grad_norm": 0.4528612196445465, | |
| "learning_rate": 9.090329093371666e-05, | |
| "loss": 0.0776, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 0.5809573531150818, | |
| "learning_rate": 8.945702546981969e-05, | |
| "loss": 0.0715, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.6470588235294117, | |
| "grad_norm": 0.5827745795249939, | |
| "learning_rate": 8.791746176467907e-05, | |
| "loss": 0.0716, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.7941176470588234, | |
| "grad_norm": 0.5849335789680481, | |
| "learning_rate": 8.628824019879137e-05, | |
| "loss": 0.0653, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.9411764705882355, | |
| "grad_norm": 0.661180317401886, | |
| "learning_rate": 8.457321315344694e-05, | |
| "loss": 0.0669, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.088235294117647, | |
| "grad_norm": 0.5251627564430237, | |
| "learning_rate": 8.277643590156894e-05, | |
| "loss": 0.069, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 3.235294117647059, | |
| "grad_norm": 0.471332311630249, | |
| "learning_rate": 8.090215701880419e-05, | |
| "loss": 0.0609, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 3.3823529411764706, | |
| "grad_norm": 0.42382729053497314, | |
| "learning_rate": 7.89548083375394e-05, | |
| "loss": 0.0622, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 3.5294117647058822, | |
| "grad_norm": 0.899319589138031, | |
| "learning_rate": 7.693899446759727e-05, | |
| "loss": 0.0648, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 3.6764705882352944, | |
| "grad_norm": 0.6775935292243958, | |
| "learning_rate": 7.485948190839077e-05, | |
| "loss": 0.0598, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 3.8235294117647056, | |
| "grad_norm": 0.43540897965431213, | |
| "learning_rate": 7.272118777828108e-05, | |
| "loss": 0.0573, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 3.9705882352941178, | |
| "grad_norm": 0.27909591794013977, | |
| "learning_rate": 7.052916818778918e-05, | |
| "loss": 0.0492, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 4.117647058823529, | |
| "grad_norm": 0.42636606097221375, | |
| "learning_rate": 6.828860628415253e-05, | |
| "loss": 0.0557, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 4.264705882352941, | |
| "grad_norm": 0.4702949821949005, | |
| "learning_rate": 6.60047999954972e-05, | |
| "loss": 0.0536, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 4.411764705882353, | |
| "grad_norm": 0.5331495404243469, | |
| "learning_rate": 6.368314950360415e-05, | |
| "loss": 0.0525, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.5588235294117645, | |
| "grad_norm": 0.301176518201828, | |
| "learning_rate": 6.132914447489137e-05, | |
| "loss": 0.0568, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 4.705882352941177, | |
| "grad_norm": 0.2303120642900467, | |
| "learning_rate": 5.8948351079804875e-05, | |
| "loss": 0.0432, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 4.852941176470588, | |
| "grad_norm": 0.37262749671936035, | |
| "learning_rate": 5.654639883131178e-05, | |
| "loss": 0.0491, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.6805188059806824, | |
| "learning_rate": 5.4128967273616625e-05, | |
| "loss": 0.0513, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 5.147058823529412, | |
| "grad_norm": 0.4015423655509949, | |
| "learning_rate": 5.170177255257618e-05, | |
| "loss": 0.0456, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 5.294117647058823, | |
| "grad_norm": 0.3756394386291504, | |
| "learning_rate": 4.9270553899567686e-05, | |
| "loss": 0.0535, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 5.4411764705882355, | |
| "grad_norm": 0.3560592532157898, | |
| "learning_rate": 4.6841060060770154e-05, | |
| "loss": 0.0463, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 5.588235294117647, | |
| "grad_norm": 0.4422471523284912, | |
| "learning_rate": 4.441903570394739e-05, | |
| "loss": 0.0417, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 5.735294117647059, | |
| "grad_norm": 0.577297568321228, | |
| "learning_rate": 4.201020783487464e-05, | |
| "loss": 0.0402, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 5.882352941176471, | |
| "grad_norm": 0.7647914290428162, | |
| "learning_rate": 3.962027225552807e-05, | |
| "loss": 0.0402, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 6.029411764705882, | |
| "grad_norm": 0.8858449459075928, | |
| "learning_rate": 3.7254880096057073e-05, | |
| "loss": 0.0479, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 6.176470588235294, | |
| "grad_norm": 0.49515625834465027, | |
| "learning_rate": 3.491962445238569e-05, | |
| "loss": 0.0434, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 6.323529411764706, | |
| "grad_norm": 0.548555850982666, | |
| "learning_rate": 3.262002716103897e-05, | |
| "loss": 0.0442, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 6.470588235294118, | |
| "grad_norm": 0.35759156942367554, | |
| "learning_rate": 3.0361525742465973e-05, | |
| "loss": 0.0411, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 6.617647058823529, | |
| "grad_norm": 0.31173354387283325, | |
| "learning_rate": 2.8149460543732664e-05, | |
| "loss": 0.0376, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 6.764705882352941, | |
| "grad_norm": 0.24947527050971985, | |
| "learning_rate": 2.598906211098643e-05, | |
| "loss": 0.0391, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 6.911764705882353, | |
| "grad_norm": 0.20982353389263153, | |
| "learning_rate": 2.388543882155067e-05, | |
| "loss": 0.0365, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 7.0588235294117645, | |
| "grad_norm": 0.3836628496646881, | |
| "learning_rate": 2.184356480489432e-05, | |
| "loss": 0.0365, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 7.205882352941177, | |
| "grad_norm": 0.23856157064437866, | |
| "learning_rate": 1.9868268181037185e-05, | |
| "loss": 0.0333, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 7.352941176470588, | |
| "grad_norm": 0.6093345880508423, | |
| "learning_rate": 1.796421964420285e-05, | |
| "loss": 0.0389, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "grad_norm": 0.2536391019821167, | |
| "learning_rate": 1.6135921418712956e-05, | |
| "loss": 0.0355, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 7.647058823529412, | |
| "grad_norm": 0.22027313709259033, | |
| "learning_rate": 1.4387696613237612e-05, | |
| "loss": 0.0331, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 7.794117647058823, | |
| "grad_norm": 0.367398738861084, | |
| "learning_rate": 1.2723678998574512e-05, | |
| "loss": 0.0352, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 7.9411764705882355, | |
| "grad_norm": 0.24603775143623352, | |
| "learning_rate": 1.114780323312724e-05, | |
| "loss": 0.0399, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 8.088235294117647, | |
| "grad_norm": 0.22743625938892365, | |
| "learning_rate": 9.663795559195733e-06, | |
| "loss": 0.0315, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 8.235294117647058, | |
| "grad_norm": 0.3211243152618408, | |
| "learning_rate": 8.275164992077556e-06, | |
| "loss": 0.033, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 8.382352941176471, | |
| "grad_norm": 0.3177715241909027, | |
| "learning_rate": 6.985195022814067e-06, | |
| "loss": 0.0366, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 8.529411764705882, | |
| "grad_norm": 0.4995149075984955, | |
| "learning_rate": 5.796935854200763e-06, | |
| "loss": 0.0353, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 8.676470588235293, | |
| "grad_norm": 0.26444506645202637, | |
| "learning_rate": 4.713197188420026e-06, | |
| "loss": 0.0311, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 8.823529411764707, | |
| "grad_norm": 0.2560294568538666, | |
| "learning_rate": 3.7365415833504725e-06, | |
| "loss": 0.0355, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 8.970588235294118, | |
| "grad_norm": 0.20665033161640167, | |
| "learning_rate": 2.869278393262226e-06, | |
| "loss": 0.0367, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 9.117647058823529, | |
| "grad_norm": 0.4208621382713318, | |
| "learning_rate": 2.113458308225458e-06, | |
| "loss": 0.0346, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 9.264705882352942, | |
| "grad_norm": 0.3151559829711914, | |
| "learning_rate": 1.4708685051444515e-06, | |
| "loss": 0.0332, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 9.411764705882353, | |
| "grad_norm": 0.5470899343490601, | |
| "learning_rate": 9.430284218824026e-07, | |
| "loss": 0.0319, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 9.558823529411764, | |
| "grad_norm": 1.0164936780929565, | |
| "learning_rate": 5.311861644696048e-07, | |
| "loss": 0.0378, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 9.705882352941176, | |
| "grad_norm": 0.18899625539779663, | |
| "learning_rate": 2.363155558901542e-07, | |
| "loss": 0.0349, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 9.852941176470589, | |
| "grad_norm": 0.18862251937389374, | |
| "learning_rate": 5.911383342556143e-08, | |
| "loss": 0.036, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.724671483039856, | |
| "learning_rate": 0.0, | |
| "loss": 0.0299, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 680, | |
| "total_flos": 9.581997557091456e+16, | |
| "train_loss": 0.08232775286716573, | |
| "train_runtime": 870.4339, | |
| "train_samples_per_second": 49.378, | |
| "train_steps_per_second": 0.781 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 680, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 10000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9.581997557091456e+16, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |