| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 61, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 3.765191996253847, | |
| "learning_rate": 2.5e-07, | |
| "loss": 1.0461, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 4.0641282279063455, | |
| "learning_rate": 5e-07, | |
| "loss": 1.0458, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 3.7765443388481783, | |
| "learning_rate": 4.996456739191904e-07, | |
| "loss": 1.0684, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 3.4137815575539796, | |
| "learning_rate": 4.985837000525343e-07, | |
| "loss": 1.0314, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 3.3692459406727107, | |
| "learning_rate": 4.968170886803361e-07, | |
| "loss": 1.0465, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 3.41903215526812, | |
| "learning_rate": 4.943508474544666e-07, | |
| "loss": 0.9973, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 3.5409185994032644, | |
| "learning_rate": 4.91191967203629e-07, | |
| "loss": 1.0003, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 3.2875389190594584, | |
| "learning_rate": 4.873494021170954e-07, | |
| "loss": 1.0126, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 3.1966679962173004, | |
| "learning_rate": 4.828340443630846e-07, | |
| "loss": 1.0149, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 3.220278106251628, | |
| "learning_rate": 4.776586932137283e-07, | |
| "loss": 1.0149, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 5.120652614407811, | |
| "learning_rate": 4.7183801876414286e-07, | |
| "loss": 1.0208, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 3.428077957669144, | |
| "learning_rate": 4.653885203484515e-07, | |
| "loss": 1.0378, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 3.7962383367663906, | |
| "learning_rate": 4.583284797706287e-07, | |
| "loss": 1.0135, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 2.9907595309256267, | |
| "learning_rate": 4.5067790948274085e-07, | |
| "loss": 0.9998, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 3.161110263982432, | |
| "learning_rate": 4.4245849585747655e-07, | |
| "loss": 1.0069, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 2.886380477987417, | |
| "learning_rate": 4.336935377157668e-07, | |
| "loss": 0.9827, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 3.376879009946129, | |
| "learning_rate": 4.244078802837462e-07, | |
| "loss": 0.9783, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 2.6892749834464516, | |
| "learning_rate": 4.146278447662597e-07, | |
| "loss": 0.9585, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 2.3786851686843105, | |
| "learning_rate": 4.0438115373654795e-07, | |
| "loss": 0.9677, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 2.1590733636497754, | |
| "learning_rate": 3.9369685255360173e-07, | |
| "loss": 0.9472, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 1.9865479320238961, | |
| "learning_rate": 3.826052270299356e-07, | |
| "loss": 0.9533, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 2.0228147756455974, | |
| "learning_rate": 3.7113771758316255e-07, | |
| "loss": 0.9366, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 1.9372326169891063, | |
| "learning_rate": 3.593268301147139e-07, | |
| "loss": 0.9316, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "grad_norm": 1.6478279831526508, | |
| "learning_rate": 3.472060438683302e-07, | |
| "loss": 0.9248, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 2.1263539873072306, | |
| "learning_rate": 3.348097165295075e-07, | |
| "loss": 0.9232, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 4.833864927411976, | |
| "learning_rate": 3.2217298683490525e-07, | |
| "loss": 0.9349, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 1.821858101250319, | |
| "learning_rate": 3.0933167496777873e-07, | |
| "loss": 0.9038, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 3.375067752782225, | |
| "learning_rate": 2.9632218102177856e-07, | |
| "loss": 0.9208, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 1.6452545248116937, | |
| "learning_rate": 2.8318138182093047e-07, | |
| "loss": 0.9091, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 1.641648446261072, | |
| "learning_rate": 2.6994652638827075e-07, | |
| "loss": 0.8863, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 1.8514627902865017, | |
| "learning_rate": 2.566551303594437e-07, | |
| "loss": 0.9176, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 1.6398219929819895, | |
| "learning_rate": 2.433448696405563e-07, | |
| "loss": 0.9132, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 1.6191504043540148, | |
| "learning_rate": 2.300534736117292e-07, | |
| "loss": 0.9188, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 1.6846506224257867, | |
| "learning_rate": 2.168186181790695e-07, | |
| "loss": 0.9207, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 2.702325993164427, | |
| "learning_rate": 2.0367781897822144e-07, | |
| "loss": 0.917, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 4.5333338294812044, | |
| "learning_rate": 1.9066832503222128e-07, | |
| "loss": 0.9267, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 1.831113061882906, | |
| "learning_rate": 1.7782701316509478e-07, | |
| "loss": 0.9067, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 1.6699336990379, | |
| "learning_rate": 1.651902834704924e-07, | |
| "loss": 0.92, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 1.8690462677412756, | |
| "learning_rate": 1.5279395613166985e-07, | |
| "loss": 0.942, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 1.9247744555110595, | |
| "learning_rate": 1.4067316988528616e-07, | |
| "loss": 0.9024, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 1.453181146995208, | |
| "learning_rate": 1.2886228241683748e-07, | |
| "loss": 0.9064, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "grad_norm": 2.049559625310238, | |
| "learning_rate": 1.173947729700644e-07, | |
| "loss": 0.9271, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 1.486378300397099, | |
| "learning_rate": 1.0630314744639829e-07, | |
| "loss": 0.9109, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 2.1416606235277196, | |
| "learning_rate": 9.561884626345204e-08, | |
| "loss": 0.904, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 1.6338101300074668, | |
| "learning_rate": 8.537215523374037e-08, | |
| "loss": 0.9148, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 2.3037793287537545, | |
| "learning_rate": 7.559211971625384e-08, | |
| "loss": 0.8772, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 1.788820498612225, | |
| "learning_rate": 6.630646228423323e-08, | |
| "loss": 0.9081, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 2.313616876377129, | |
| "learning_rate": 5.75415041425234e-08, | |
| "loss": 0.93, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 1.8442253511582427, | |
| "learning_rate": 4.932209051725914e-08, | |
| "loss": 0.9303, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 1.797668108469576, | |
| "learning_rate": 4.167152022937123e-08, | |
| "loss": 0.9046, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 2.0071094870512707, | |
| "learning_rate": 3.461147965154845e-08, | |
| "loss": 0.8853, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 1.5091131315393522, | |
| "learning_rate": 2.816198123585714e-08, | |
| "loss": 0.893, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 2.291025373137871, | |
| "learning_rate": 2.234130678627169e-08, | |
| "loss": 0.9102, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 1.975139175162289, | |
| "learning_rate": 1.7165955636915392e-08, | |
| "loss": 0.9448, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 1.9662365790545857, | |
| "learning_rate": 1.265059788290468e-08, | |
| "loss": 0.9126, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 1.5860718910737048, | |
| "learning_rate": 8.808032796371017e-09, | |
| "loss": 0.9008, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 5.824152576947844, | |
| "learning_rate": 5.649152545533331e-09, | |
| "loss": 0.9104, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 1.8084725184982782, | |
| "learning_rate": 3.1829113196638613e-09, | |
| "loss": 0.9108, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 2.4499467355310878, | |
| "learning_rate": 1.4162999474657266e-09, | |
| "loss": 0.9218, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "grad_norm": 1.7234130969562493, | |
| "learning_rate": 3.5432608080951386e-10, | |
| "loss": 0.913, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.5468535844514548, | |
| "learning_rate": 0.0, | |
| "loss": 0.9164, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 61, | |
| "total_flos": 57696792150016.0, | |
| "train_loss": 0.9464570713824914, | |
| "train_runtime": 1092.378, | |
| "train_samples_per_second": 14.334, | |
| "train_steps_per_second": 0.056 | |
| } | |
| ], | |
| "logging_steps": 1.0, | |
| "max_steps": 61, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 1000, | |
| "total_flos": 57696792150016.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |