| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.72, | |
| "eval_steps": 500, | |
| "global_step": 70, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.096, | |
| "grad_norm": 6.663326141730961, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 1.0516, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 6.796065411621461, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 1.0783, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 6.5933447825542295, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 1.1187, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 2.360990134630162, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.9758, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 5.13190132899842, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.9871, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 5.627747980993312, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.9494, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 5.17633351595843, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9427, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 3.9253036943236297, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.9337, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 4.2350002586318976, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.8716, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 3.2195649485039315, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.9377, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 4.31233016378484, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 1.3488, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 2.64965167837867, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.824, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 1.7081096486050087, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.817, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 1.439084150798308, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.7187, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 1.1368635392554118, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.8003, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 1.2070533921577575, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.7207, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 1.1813304161965994, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.8002, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 0.7172373337868946, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.6485, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 0.9366203299867314, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.7291, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.9205161586181824, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.7655, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 1.9624387434073385, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 1.1319, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 0.9154546033857803, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.5751, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 0.8708675192048209, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.6297, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 1.1297839514640866, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.618, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.8845469763148248, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.5573, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 2.975438538360796, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.6124, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 1.518519382460017, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.5849, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 2.771601710934844, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.5443, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 0.8165057217338433, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.6143, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.7159068051012769, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.5837, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 0.9385300200934704, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.6278, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.072, | |
| "grad_norm": 3.686721838150586, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.6724, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.168, | |
| "grad_norm": 0.8068292457096653, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.4396, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.2640000000000002, | |
| "grad_norm": 0.764096501155502, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.4402, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.7606514567009977, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.465, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.456, | |
| "grad_norm": 0.6808959563785313, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.4295, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.552, | |
| "grad_norm": 0.682898251817044, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.4696, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.648, | |
| "grad_norm": 0.5651140159558662, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.4144, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7439999999999998, | |
| "grad_norm": 0.6886978866991732, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.4752, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.5717398825196983, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.416, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.936, | |
| "grad_norm": 0.5937242464554575, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.4446, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.032, | |
| "grad_norm": 1.0330731883429842, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.6667, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.128, | |
| "grad_norm": 0.6508087375958522, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.3602, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.224, | |
| "grad_norm": 0.6360404008558569, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.3691, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.5231316278881759, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.3129, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.416, | |
| "grad_norm": 0.7568156362330419, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.3357, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.5120000000000005, | |
| "grad_norm": 0.7400387529584419, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.3358, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.608, | |
| "grad_norm": 0.5662611289093206, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.3421, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.704, | |
| "grad_norm": 0.5564581474877931, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.2855, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.6051844055416135, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.3353, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.896, | |
| "grad_norm": 0.5353232131242458, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.2999, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.992, | |
| "grad_norm": 0.8068722052327139, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.5013, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.088, | |
| "grad_norm": 0.6427093848572188, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.2869, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.184, | |
| "grad_norm": 0.5118821520176856, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.2787, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 0.506625378500426, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.2716, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.376, | |
| "grad_norm": 0.5566638158414696, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.2618, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.4719999999999995, | |
| "grad_norm": 0.5708909474894296, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.2674, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.568, | |
| "grad_norm": 0.5407929355519842, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.2567, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.664, | |
| "grad_norm": 0.47296368483911794, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.2614, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 0.4745650664247802, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.2744, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.856, | |
| "grad_norm": 0.4880392998094458, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.2547, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.952, | |
| "grad_norm": 0.48726675161470534, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.2689, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.048, | |
| "grad_norm": 0.8983955517219776, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.4059, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.144, | |
| "grad_norm": 0.5437514089476584, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.2622, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 0.4549325884342243, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.2198, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.336, | |
| "grad_norm": 0.46162765738382716, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.2578, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.432, | |
| "grad_norm": 0.4562447234789301, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.2446, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.5280000000000005, | |
| "grad_norm": 0.4820361478151564, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.2475, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.624, | |
| "grad_norm": 0.41143341313807275, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.2173, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.466836980358744, | |
| "learning_rate": 0.0, | |
| "loss": 0.2461, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "step": 70, | |
| "total_flos": 8.072515081889382e+16, | |
| "train_loss": 0.5527579554489681, | |
| "train_runtime": 3250.7929, | |
| "train_samples_per_second": 2.153, | |
| "train_steps_per_second": 0.022 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 70, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 8.072515081889382e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |