|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 0.9900199600798403, |
|
"eval_steps": 500, |
|
"global_step": 62, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.02, |
|
"grad_norm": 34.680301666259766, |
|
"learning_rate": 1e-05, |
|
"loss": 16.3759, |
|
"step": 1 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"grad_norm": 3.369790554046631, |
|
"learning_rate": 1e-05, |
|
"loss": 0.6651, |
|
"step": 2 |
|
}, |
|
{ |
|
"epoch": 0.05, |
|
"grad_norm": 3.2419369220733643, |
|
"learning_rate": 1e-05, |
|
"loss": 0.5262, |
|
"step": 3 |
|
}, |
|
{ |
|
"epoch": 0.06, |
|
"grad_norm": 2.575935125350952, |
|
"learning_rate": 1e-05, |
|
"loss": 0.446, |
|
"step": 4 |
|
}, |
|
{ |
|
"epoch": 0.08, |
|
"grad_norm": 2.4852099418640137, |
|
"learning_rate": 1e-05, |
|
"loss": 0.3491, |
|
"step": 5 |
|
}, |
|
{ |
|
"epoch": 0.1, |
|
"grad_norm": 0.8563106060028076, |
|
"learning_rate": 1e-05, |
|
"loss": 0.3045, |
|
"step": 6 |
|
}, |
|
{ |
|
"epoch": 0.11, |
|
"grad_norm": 4.232894420623779, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2871, |
|
"step": 7 |
|
}, |
|
{ |
|
"epoch": 0.13, |
|
"grad_norm": 0.6341336965560913, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2612, |
|
"step": 8 |
|
}, |
|
{ |
|
"epoch": 0.14, |
|
"grad_norm": 0.8098199963569641, |
|
"learning_rate": 1e-05, |
|
"loss": 591.0046, |
|
"step": 9 |
|
}, |
|
{ |
|
"epoch": 0.16, |
|
"grad_norm": 0.407880961894989, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2701, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.18, |
|
"grad_norm": 0.34948238730430603, |
|
"learning_rate": 1e-05, |
|
"loss": 0.277, |
|
"step": 11 |
|
}, |
|
{ |
|
"epoch": 0.19, |
|
"grad_norm": 0.3844144642353058, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2608, |
|
"step": 12 |
|
}, |
|
{ |
|
"epoch": 0.21, |
|
"grad_norm": 0.39868655800819397, |
|
"learning_rate": 1e-05, |
|
"loss": 0.257, |
|
"step": 13 |
|
}, |
|
{ |
|
"epoch": 0.22, |
|
"grad_norm": 0.3178333640098572, |
|
"learning_rate": 1e-05, |
|
"loss": 0.242, |
|
"step": 14 |
|
}, |
|
{ |
|
"epoch": 0.24, |
|
"grad_norm": 0.28359681367874146, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2391, |
|
"step": 15 |
|
}, |
|
{ |
|
"epoch": 0.26, |
|
"grad_norm": 0.3104497790336609, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2297, |
|
"step": 16 |
|
}, |
|
{ |
|
"epoch": 0.27, |
|
"grad_norm": 0.7799801826477051, |
|
"learning_rate": 1e-05, |
|
"loss": 81.355, |
|
"step": 17 |
|
}, |
|
{ |
|
"epoch": 0.29, |
|
"grad_norm": 0.2582087218761444, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2603, |
|
"step": 18 |
|
}, |
|
{ |
|
"epoch": 0.3, |
|
"grad_norm": 0.2229820191860199, |
|
"learning_rate": 1e-05, |
|
"loss": 0.255, |
|
"step": 19 |
|
}, |
|
{ |
|
"epoch": 0.32, |
|
"grad_norm": 0.2470751404762268, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2487, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.34, |
|
"grad_norm": 0.28791287541389465, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2408, |
|
"step": 21 |
|
}, |
|
{ |
|
"epoch": 0.35, |
|
"grad_norm": 0.2907260060310364, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2396, |
|
"step": 22 |
|
}, |
|
{ |
|
"epoch": 0.37, |
|
"grad_norm": 0.2796117663383484, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2281, |
|
"step": 23 |
|
}, |
|
{ |
|
"epoch": 0.38, |
|
"grad_norm": 0.3333979547023773, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2261, |
|
"step": 24 |
|
}, |
|
{ |
|
"epoch": 0.4, |
|
"grad_norm": 0.4846266508102417, |
|
"learning_rate": 1e-05, |
|
"loss": 63.2997, |
|
"step": 25 |
|
}, |
|
{ |
|
"epoch": 0.42, |
|
"grad_norm": 0.20356160402297974, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2481, |
|
"step": 26 |
|
}, |
|
{ |
|
"epoch": 0.43, |
|
"grad_norm": 0.2213103324174881, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2387, |
|
"step": 27 |
|
}, |
|
{ |
|
"epoch": 0.45, |
|
"grad_norm": 0.26175907254219055, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2475, |
|
"step": 28 |
|
}, |
|
{ |
|
"epoch": 0.46, |
|
"grad_norm": 0.2190675288438797, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2379, |
|
"step": 29 |
|
}, |
|
{ |
|
"epoch": 0.48, |
|
"grad_norm": 0.24093040823936462, |
|
"learning_rate": 1e-05, |
|
"loss": 0.239, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.5, |
|
"grad_norm": 0.277065247297287, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2362, |
|
"step": 31 |
|
}, |
|
{ |
|
"epoch": 0.51, |
|
"grad_norm": 0.33813706040382385, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2174, |
|
"step": 32 |
|
}, |
|
{ |
|
"epoch": 0.53, |
|
"grad_norm": 0.40611541271209717, |
|
"learning_rate": 1e-05, |
|
"loss": 203.4452, |
|
"step": 33 |
|
}, |
|
{ |
|
"epoch": 0.54, |
|
"grad_norm": 0.2279174029827118, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2388, |
|
"step": 34 |
|
}, |
|
{ |
|
"epoch": 0.56, |
|
"grad_norm": 0.21085476875305176, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2417, |
|
"step": 35 |
|
}, |
|
{ |
|
"epoch": 0.57, |
|
"grad_norm": 0.2684917151927948, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2426, |
|
"step": 36 |
|
}, |
|
{ |
|
"epoch": 0.59, |
|
"grad_norm": 0.24274484813213348, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2409, |
|
"step": 37 |
|
}, |
|
{ |
|
"epoch": 0.61, |
|
"grad_norm": 0.26018860936164856, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2212, |
|
"step": 38 |
|
}, |
|
{ |
|
"epoch": 0.62, |
|
"grad_norm": 0.2579266130924225, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2281, |
|
"step": 39 |
|
}, |
|
{ |
|
"epoch": 0.64, |
|
"grad_norm": 0.31496453285217285, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2149, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.65, |
|
"grad_norm": 0.4951270520687103, |
|
"learning_rate": 1e-05, |
|
"loss": 126.4606, |
|
"step": 41 |
|
}, |
|
{ |
|
"epoch": 0.67, |
|
"grad_norm": 0.2128014862537384, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2417, |
|
"step": 42 |
|
}, |
|
{ |
|
"epoch": 0.69, |
|
"grad_norm": 0.23777072131633759, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2327, |
|
"step": 43 |
|
}, |
|
{ |
|
"epoch": 0.7, |
|
"grad_norm": 0.20361731946468353, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2301, |
|
"step": 44 |
|
}, |
|
{ |
|
"epoch": 0.72, |
|
"grad_norm": 0.2542952001094818, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2406, |
|
"step": 45 |
|
}, |
|
{ |
|
"epoch": 0.73, |
|
"grad_norm": 0.283297598361969, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2328, |
|
"step": 46 |
|
}, |
|
{ |
|
"epoch": 0.75, |
|
"grad_norm": 0.261777400970459, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2246, |
|
"step": 47 |
|
}, |
|
{ |
|
"epoch": 0.77, |
|
"grad_norm": 0.27042126655578613, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2136, |
|
"step": 48 |
|
}, |
|
{ |
|
"epoch": 0.78, |
|
"grad_norm": 0.45323169231414795, |
|
"learning_rate": 1e-05, |
|
"loss": 37.1518, |
|
"step": 49 |
|
}, |
|
{ |
|
"epoch": 0.8, |
|
"grad_norm": 0.19414253532886505, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2423, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.81, |
|
"grad_norm": 0.2136753797531128, |
|
"learning_rate": 1e-05, |
|
"loss": 0.237, |
|
"step": 51 |
|
}, |
|
{ |
|
"epoch": 0.83, |
|
"grad_norm": 0.23989234864711761, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2348, |
|
"step": 52 |
|
}, |
|
{ |
|
"epoch": 0.85, |
|
"grad_norm": 0.23847776651382446, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2357, |
|
"step": 53 |
|
}, |
|
{ |
|
"epoch": 0.86, |
|
"grad_norm": 0.2265355885028839, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2276, |
|
"step": 54 |
|
}, |
|
{ |
|
"epoch": 0.88, |
|
"grad_norm": 0.2720184326171875, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2215, |
|
"step": 55 |
|
}, |
|
{ |
|
"epoch": 0.89, |
|
"grad_norm": 0.32160285115242004, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2123, |
|
"step": 56 |
|
}, |
|
{ |
|
"epoch": 0.91, |
|
"grad_norm": 0.4474925100803375, |
|
"learning_rate": 1e-05, |
|
"loss": 82.7252, |
|
"step": 57 |
|
}, |
|
{ |
|
"epoch": 0.93, |
|
"grad_norm": 0.22881349921226501, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2473, |
|
"step": 58 |
|
}, |
|
{ |
|
"epoch": 0.94, |
|
"grad_norm": 0.2213418185710907, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2329, |
|
"step": 59 |
|
}, |
|
{ |
|
"epoch": 0.96, |
|
"grad_norm": 0.23734869062900543, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2313, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 0.97, |
|
"grad_norm": 0.26081162691116333, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2365, |
|
"step": 61 |
|
}, |
|
{ |
|
"epoch": 0.99, |
|
"grad_norm": 0.29037076234817505, |
|
"learning_rate": 1e-05, |
|
"loss": 0.2291, |
|
"step": 62 |
|
}, |
|
{ |
|
"epoch": 0.99, |
|
"step": 62, |
|
"total_flos": 190914982051840.0, |
|
"train_loss": 19.610142173545977, |
|
"train_runtime": 47287.2785, |
|
"train_samples_per_second": 0.021, |
|
"train_steps_per_second": 0.001 |
|
} |
|
], |
|
"logging_steps": 1.0, |
|
"max_steps": 62, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 1, |
|
"save_steps": 20, |
|
"total_flos": 190914982051840.0, |
|
"train_batch_size": 1, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|