| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9448818897637795, | |
| "eval_steps": 5, | |
| "global_step": 60, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.015748031496062992, | |
| "grad_norm": 5.510926246643066, | |
| "learning_rate": 1.9894179894179895e-05, | |
| "loss": 0.9249, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.031496062992125984, | |
| "grad_norm": 8.61505126953125, | |
| "learning_rate": 1.978835978835979e-05, | |
| "loss": 0.8445, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.047244094488188976, | |
| "grad_norm": 7.036591529846191, | |
| "learning_rate": 1.9682539682539684e-05, | |
| "loss": 0.9654, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.06299212598425197, | |
| "grad_norm": 5.803933143615723, | |
| "learning_rate": 1.9576719576719577e-05, | |
| "loss": 0.9276, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.07874015748031496, | |
| "grad_norm": 5.716428756713867, | |
| "learning_rate": 1.947089947089947e-05, | |
| "loss": 0.9241, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.07874015748031496, | |
| "eval_accuracy": 0.5678496868475992, | |
| "eval_loss": 0.6996241807937622, | |
| "eval_runtime": 108.7291, | |
| "eval_samples_per_second": 4.405, | |
| "eval_steps_per_second": 0.552, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.09448818897637795, | |
| "grad_norm": 7.55866813659668, | |
| "learning_rate": 1.9365079365079367e-05, | |
| "loss": 0.9947, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.11023622047244094, | |
| "grad_norm": 6.801171779632568, | |
| "learning_rate": 1.925925925925926e-05, | |
| "loss": 0.972, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.12598425196850394, | |
| "grad_norm": 4.845946311950684, | |
| "learning_rate": 1.9153439153439156e-05, | |
| "loss": 0.6478, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.14173228346456693, | |
| "grad_norm": 10.487945556640625, | |
| "learning_rate": 1.904761904761905e-05, | |
| "loss": 0.8597, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.15748031496062992, | |
| "grad_norm": 5.452786445617676, | |
| "learning_rate": 1.8941798941798943e-05, | |
| "loss": 0.7708, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.15748031496062992, | |
| "eval_accuracy": 0.6659707724425887, | |
| "eval_loss": 0.6283570528030396, | |
| "eval_runtime": 108.7155, | |
| "eval_samples_per_second": 4.406, | |
| "eval_steps_per_second": 0.552, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.1732283464566929, | |
| "grad_norm": 4.522532939910889, | |
| "learning_rate": 1.8835978835978836e-05, | |
| "loss": 0.6848, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.1889763779527559, | |
| "grad_norm": 6.4987688064575195, | |
| "learning_rate": 1.8730158730158732e-05, | |
| "loss": 0.6644, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.2047244094488189, | |
| "grad_norm": 4.2297682762146, | |
| "learning_rate": 1.8624338624338625e-05, | |
| "loss": 0.7227, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.2204724409448819, | |
| "grad_norm": 6.5658063888549805, | |
| "learning_rate": 1.851851851851852e-05, | |
| "loss": 0.6991, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.23622047244094488, | |
| "grad_norm": 6.549685001373291, | |
| "learning_rate": 1.8412698412698415e-05, | |
| "loss": 0.7875, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.23622047244094488, | |
| "eval_accuracy": 0.7244258872651357, | |
| "eval_loss": 0.5749094486236572, | |
| "eval_runtime": 108.7121, | |
| "eval_samples_per_second": 4.406, | |
| "eval_steps_per_second": 0.552, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.25196850393700787, | |
| "grad_norm": 3.6349198818206787, | |
| "learning_rate": 1.8306878306878308e-05, | |
| "loss": 0.5732, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.2677165354330709, | |
| "grad_norm": 4.741979598999023, | |
| "learning_rate": 1.82010582010582e-05, | |
| "loss": 0.5774, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.28346456692913385, | |
| "grad_norm": 4.751223087310791, | |
| "learning_rate": 1.8095238095238097e-05, | |
| "loss": 0.5738, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.2992125984251969, | |
| "grad_norm": 5.214819431304932, | |
| "learning_rate": 1.798941798941799e-05, | |
| "loss": 0.7182, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.31496062992125984, | |
| "grad_norm": 5.566962718963623, | |
| "learning_rate": 1.7883597883597884e-05, | |
| "loss": 0.6575, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.31496062992125984, | |
| "eval_accuracy": 0.7390396659707724, | |
| "eval_loss": 0.5360159873962402, | |
| "eval_runtime": 108.7252, | |
| "eval_samples_per_second": 4.406, | |
| "eval_steps_per_second": 0.552, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.33070866141732286, | |
| "grad_norm": 4.060683727264404, | |
| "learning_rate": 1.7777777777777777e-05, | |
| "loss": 0.5976, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.3464566929133858, | |
| "grad_norm": 5.9868621826171875, | |
| "learning_rate": 1.7671957671957673e-05, | |
| "loss": 0.7734, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.36220472440944884, | |
| "grad_norm": 3.4295496940612793, | |
| "learning_rate": 1.7566137566137566e-05, | |
| "loss": 0.5543, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.3779527559055118, | |
| "grad_norm": 4.587719917297363, | |
| "learning_rate": 1.7460317460317463e-05, | |
| "loss": 0.6497, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.3937007874015748, | |
| "grad_norm": 4.520890235900879, | |
| "learning_rate": 1.7354497354497356e-05, | |
| "loss": 0.6802, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.3937007874015748, | |
| "eval_accuracy": 0.7432150313152401, | |
| "eval_loss": 0.5086582899093628, | |
| "eval_runtime": 108.7194, | |
| "eval_samples_per_second": 4.406, | |
| "eval_steps_per_second": 0.552, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.4094488188976378, | |
| "grad_norm": 3.653116464614868, | |
| "learning_rate": 1.724867724867725e-05, | |
| "loss": 0.4888, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.4251968503937008, | |
| "grad_norm": 4.042315483093262, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.6004, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.4409448818897638, | |
| "grad_norm": 5.317520618438721, | |
| "learning_rate": 1.7037037037037038e-05, | |
| "loss": 0.6253, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.4566929133858268, | |
| "grad_norm": 3.8642020225524902, | |
| "learning_rate": 1.693121693121693e-05, | |
| "loss": 0.5778, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.47244094488188976, | |
| "grad_norm": 2.3941361904144287, | |
| "learning_rate": 1.6825396825396828e-05, | |
| "loss": 0.3982, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.47244094488188976, | |
| "eval_accuracy": 0.7578288100208769, | |
| "eval_loss": 0.4889708459377289, | |
| "eval_runtime": 108.7183, | |
| "eval_samples_per_second": 4.406, | |
| "eval_steps_per_second": 0.552, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.4881889763779528, | |
| "grad_norm": 4.1248650550842285, | |
| "learning_rate": 1.671957671957672e-05, | |
| "loss": 0.777, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.5039370078740157, | |
| "grad_norm": 3.369483470916748, | |
| "learning_rate": 1.6613756613756614e-05, | |
| "loss": 0.5675, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.5196850393700787, | |
| "grad_norm": 3.8457119464874268, | |
| "learning_rate": 1.6507936507936507e-05, | |
| "loss": 0.6227, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.5354330708661418, | |
| "grad_norm": 4.809354782104492, | |
| "learning_rate": 1.6402116402116404e-05, | |
| "loss": 0.7111, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.5511811023622047, | |
| "grad_norm": 2.84769868850708, | |
| "learning_rate": 1.6296296296296297e-05, | |
| "loss": 0.4555, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.5511811023622047, | |
| "eval_accuracy": 0.7599164926931107, | |
| "eval_loss": 0.4774630665779114, | |
| "eval_runtime": 108.7145, | |
| "eval_samples_per_second": 4.406, | |
| "eval_steps_per_second": 0.552, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.5669291338582677, | |
| "grad_norm": 4.496406555175781, | |
| "learning_rate": 1.6190476190476193e-05, | |
| "loss": 0.6703, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.5826771653543307, | |
| "grad_norm": 5.721245288848877, | |
| "learning_rate": 1.6084656084656086e-05, | |
| "loss": 0.7066, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.5984251968503937, | |
| "grad_norm": 4.494580268859863, | |
| "learning_rate": 1.597883597883598e-05, | |
| "loss": 0.4907, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.6141732283464567, | |
| "grad_norm": 2.8905560970306396, | |
| "learning_rate": 1.5873015873015872e-05, | |
| "loss": 0.5501, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.6299212598425197, | |
| "grad_norm": 9.776362419128418, | |
| "learning_rate": 1.576719576719577e-05, | |
| "loss": 0.8838, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.6299212598425197, | |
| "eval_accuracy": 0.7661795407098121, | |
| "eval_loss": 0.46829721331596375, | |
| "eval_runtime": 108.7189, | |
| "eval_samples_per_second": 4.406, | |
| "eval_steps_per_second": 0.552, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.6456692913385826, | |
| "grad_norm": 3.8481881618499756, | |
| "learning_rate": 1.5661375661375662e-05, | |
| "loss": 0.5309, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.6614173228346457, | |
| "grad_norm": 6.0327839851379395, | |
| "learning_rate": 1.555555555555556e-05, | |
| "loss": 0.6414, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.6771653543307087, | |
| "grad_norm": 4.993657112121582, | |
| "learning_rate": 1.544973544973545e-05, | |
| "loss": 0.5727, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.6929133858267716, | |
| "grad_norm": 4.3265252113342285, | |
| "learning_rate": 1.5343915343915344e-05, | |
| "loss": 0.4913, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.7086614173228346, | |
| "grad_norm": 3.6012353897094727, | |
| "learning_rate": 1.523809523809524e-05, | |
| "loss": 0.4692, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.7086614173228346, | |
| "eval_accuracy": 0.7661795407098121, | |
| "eval_loss": 0.4610559344291687, | |
| "eval_runtime": 108.7229, | |
| "eval_samples_per_second": 4.406, | |
| "eval_steps_per_second": 0.552, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.7244094488188977, | |
| "grad_norm": 4.319406509399414, | |
| "learning_rate": 1.5132275132275134e-05, | |
| "loss": 0.5203, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.7401574803149606, | |
| "grad_norm": 3.885263442993164, | |
| "learning_rate": 1.5026455026455027e-05, | |
| "loss": 0.5084, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.7559055118110236, | |
| "grad_norm": 3.547327995300293, | |
| "learning_rate": 1.4920634920634922e-05, | |
| "loss": 0.442, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.7716535433070866, | |
| "grad_norm": 3.8868982791900635, | |
| "learning_rate": 1.4814814814814815e-05, | |
| "loss": 0.5848, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.7874015748031497, | |
| "grad_norm": 2.222346544265747, | |
| "learning_rate": 1.470899470899471e-05, | |
| "loss": 0.5455, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.7874015748031497, | |
| "eval_accuracy": 0.7620041753653445, | |
| "eval_loss": 0.4531377851963043, | |
| "eval_runtime": 108.7528, | |
| "eval_samples_per_second": 4.404, | |
| "eval_steps_per_second": 0.552, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.8031496062992126, | |
| "grad_norm": 3.129575252532959, | |
| "learning_rate": 1.4603174603174603e-05, | |
| "loss": 0.4861, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.8188976377952756, | |
| "grad_norm": 4.924710750579834, | |
| "learning_rate": 1.44973544973545e-05, | |
| "loss": 0.5782, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.8346456692913385, | |
| "grad_norm": 5.2157182693481445, | |
| "learning_rate": 1.4391534391534392e-05, | |
| "loss": 0.7203, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.8503937007874016, | |
| "grad_norm": 4.697371959686279, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.4261, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.8661417322834646, | |
| "grad_norm": 2.8899056911468506, | |
| "learning_rate": 1.417989417989418e-05, | |
| "loss": 0.5696, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.8661417322834646, | |
| "eval_accuracy": 0.7661795407098121, | |
| "eval_loss": 0.4459321200847626, | |
| "eval_runtime": 108.7951, | |
| "eval_samples_per_second": 4.403, | |
| "eval_steps_per_second": 0.551, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.8818897637795275, | |
| "grad_norm": 4.532041072845459, | |
| "learning_rate": 1.4074074074074075e-05, | |
| "loss": 0.5723, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.8976377952755905, | |
| "grad_norm": 2.3436343669891357, | |
| "learning_rate": 1.3968253968253968e-05, | |
| "loss": 0.3629, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.9133858267716536, | |
| "grad_norm": 3.333158493041992, | |
| "learning_rate": 1.3862433862433865e-05, | |
| "loss": 0.5433, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.9291338582677166, | |
| "grad_norm": 4.177884101867676, | |
| "learning_rate": 1.3756613756613758e-05, | |
| "loss": 0.3747, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.9448818897637795, | |
| "grad_norm": 5.238712310791016, | |
| "learning_rate": 1.3650793650793652e-05, | |
| "loss": 0.7453, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.9448818897637795, | |
| "eval_accuracy": 0.7766179540709812, | |
| "eval_loss": 0.4413756728172302, | |
| "eval_runtime": 108.7463, | |
| "eval_samples_per_second": 4.405, | |
| "eval_steps_per_second": 0.552, | |
| "step": 60 | |
| } | |
| ], | |
| "logging_steps": 1.0, | |
| "max_steps": 189, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 20, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |