| { | |
| "best_metric": 0.8683206106870229, | |
| "best_model_checkpoint": "beit-base-patch16-224-pt22k-ft22k-finetuned-lora-medmnistv2/checkpoint-590", | |
| "epoch": 9.898305084745763, | |
| "eval_steps": 500, | |
| "global_step": 730, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.13559322033898305, | |
| "grad_norm": 12.995793342590332, | |
| "learning_rate": 0.004945205479452055, | |
| "loss": 0.7148, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.2711864406779661, | |
| "grad_norm": 1.2968316078186035, | |
| "learning_rate": 0.004876712328767123, | |
| "loss": 0.612, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.4067796610169492, | |
| "grad_norm": 1.6618480682373047, | |
| "learning_rate": 0.004808219178082192, | |
| "loss": 0.5877, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.5423728813559322, | |
| "grad_norm": 5.746555328369141, | |
| "learning_rate": 0.00473972602739726, | |
| "loss": 0.5768, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.6779661016949152, | |
| "grad_norm": 1.3945930004119873, | |
| "learning_rate": 0.0046712328767123295, | |
| "loss": 0.5698, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.8135593220338984, | |
| "grad_norm": 6.072067737579346, | |
| "learning_rate": 0.004602739726027398, | |
| "loss": 0.5479, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.9491525423728814, | |
| "grad_norm": 0.36280161142349243, | |
| "learning_rate": 0.004534246575342466, | |
| "loss": 0.5947, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.9898305084745763, | |
| "eval_accuracy": 0.7423664122137404, | |
| "eval_f1": 0.42606790799561883, | |
| "eval_loss": 0.5164713263511658, | |
| "eval_precision": 0.3711832061068702, | |
| "eval_recall": 0.5, | |
| "eval_runtime": 2.9766, | |
| "eval_samples_per_second": 176.038, | |
| "eval_steps_per_second": 11.086, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 1.0847457627118644, | |
| "grad_norm": 3.3023428916931152, | |
| "learning_rate": 0.004465753424657534, | |
| "loss": 0.5075, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.2203389830508475, | |
| "grad_norm": 4.128161430358887, | |
| "learning_rate": 0.004397260273972603, | |
| "loss": 0.641, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.3559322033898304, | |
| "grad_norm": 7.110351085662842, | |
| "learning_rate": 0.004328767123287672, | |
| "loss": 0.7467, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.4915254237288136, | |
| "grad_norm": 0.7752661108970642, | |
| "learning_rate": 0.00426027397260274, | |
| "loss": 0.5226, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.6271186440677967, | |
| "grad_norm": 0.3319437801837921, | |
| "learning_rate": 0.004191780821917808, | |
| "loss": 0.5337, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.7627118644067796, | |
| "grad_norm": 1.7883654832839966, | |
| "learning_rate": 0.004123287671232877, | |
| "loss": 0.4819, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.8983050847457628, | |
| "grad_norm": 0.5220211148262024, | |
| "learning_rate": 0.0040547945205479455, | |
| "loss": 0.4888, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.993220338983051, | |
| "eval_accuracy": 0.8568702290076335, | |
| "eval_f1": 0.8151240738562859, | |
| "eval_loss": 0.3449835777282715, | |
| "eval_precision": 0.8115699404761905, | |
| "eval_recall": 0.8189564886222984, | |
| "eval_runtime": 3.0609, | |
| "eval_samples_per_second": 171.191, | |
| "eval_steps_per_second": 10.781, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 2.0338983050847457, | |
| "grad_norm": 5.3052544593811035, | |
| "learning_rate": 0.003986301369863014, | |
| "loss": 0.4278, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.169491525423729, | |
| "grad_norm": 4.525509357452393, | |
| "learning_rate": 0.003917808219178082, | |
| "loss": 0.4448, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.305084745762712, | |
| "grad_norm": 6.519794940948486, | |
| "learning_rate": 0.0038493150684931507, | |
| "loss": 0.4817, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.440677966101695, | |
| "grad_norm": 1.3314294815063477, | |
| "learning_rate": 0.0037808219178082194, | |
| "loss": 0.4393, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.576271186440678, | |
| "grad_norm": 0.6930696368217468, | |
| "learning_rate": 0.0037123287671232876, | |
| "loss": 0.3842, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.711864406779661, | |
| "grad_norm": 3.826169729232788, | |
| "learning_rate": 0.0036438356164383563, | |
| "loss": 0.4206, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.847457627118644, | |
| "grad_norm": 0.5641322135925293, | |
| "learning_rate": 0.0035753424657534246, | |
| "loss": 0.3684, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.983050847457627, | |
| "grad_norm": 3.910162925720215, | |
| "learning_rate": 0.0035068493150684932, | |
| "loss": 0.4022, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.9966101694915253, | |
| "eval_accuracy": 0.833969465648855, | |
| "eval_f1": 0.8079106043560151, | |
| "eval_loss": 0.42247942090034485, | |
| "eval_precision": 0.7914074166251867, | |
| "eval_recall": 0.8567361706179186, | |
| "eval_runtime": 3.2139, | |
| "eval_samples_per_second": 163.043, | |
| "eval_steps_per_second": 10.268, | |
| "step": 221 | |
| }, | |
| { | |
| "epoch": 3.1186440677966103, | |
| "grad_norm": 1.5272403955459595, | |
| "learning_rate": 0.0034383561643835615, | |
| "loss": 0.4376, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 3.2542372881355934, | |
| "grad_norm": 0.9074549674987793, | |
| "learning_rate": 0.00336986301369863, | |
| "loss": 0.403, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 3.389830508474576, | |
| "grad_norm": 0.9038354158401489, | |
| "learning_rate": 0.0033013698630136984, | |
| "loss": 0.3898, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 3.5254237288135593, | |
| "grad_norm": 0.5530111193656921, | |
| "learning_rate": 0.003232876712328767, | |
| "loss": 0.3958, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 3.6610169491525424, | |
| "grad_norm": 3.4515538215637207, | |
| "learning_rate": 0.0031643835616438354, | |
| "loss": 0.4354, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 3.7966101694915255, | |
| "grad_norm": 5.728641986846924, | |
| "learning_rate": 0.003095890410958904, | |
| "loss": 0.4601, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 3.9322033898305087, | |
| "grad_norm": 0.6854091882705688, | |
| "learning_rate": 0.0030273972602739723, | |
| "loss": 0.4319, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8587786259541985, | |
| "eval_f1": 0.8292197382097493, | |
| "eval_loss": 0.3599797189235687, | |
| "eval_precision": 0.8122691052465915, | |
| "eval_recall": 0.8589355422260307, | |
| "eval_runtime": 2.983, | |
| "eval_samples_per_second": 175.665, | |
| "eval_steps_per_second": 11.063, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 4.067796610169491, | |
| "grad_norm": 4.648618221282959, | |
| "learning_rate": 0.002958904109589041, | |
| "loss": 0.3716, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.203389830508475, | |
| "grad_norm": 2.3614566326141357, | |
| "learning_rate": 0.0028904109589041093, | |
| "loss": 0.423, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 4.338983050847458, | |
| "grad_norm": 1.8763720989227295, | |
| "learning_rate": 0.0028219178082191784, | |
| "loss": 0.3737, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 4.47457627118644, | |
| "grad_norm": 3.2308576107025146, | |
| "learning_rate": 0.002753424657534246, | |
| "loss": 0.4613, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 4.610169491525424, | |
| "grad_norm": 1.1305418014526367, | |
| "learning_rate": 0.0026849315068493153, | |
| "loss": 0.377, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 4.745762711864407, | |
| "grad_norm": 0.8450165390968323, | |
| "learning_rate": 0.0026164383561643836, | |
| "loss": 0.3981, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 4.88135593220339, | |
| "grad_norm": 0.7082638144493103, | |
| "learning_rate": 0.0025479452054794523, | |
| "loss": 0.3836, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 4.989830508474577, | |
| "eval_accuracy": 0.851145038167939, | |
| "eval_f1": 0.8233035899564225, | |
| "eval_loss": 0.36647942662239075, | |
| "eval_precision": 0.8054056666076059, | |
| "eval_recall": 0.8610492240312292, | |
| "eval_runtime": 3.005, | |
| "eval_samples_per_second": 174.375, | |
| "eval_steps_per_second": 10.982, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 5.016949152542373, | |
| "grad_norm": 0.8164511322975159, | |
| "learning_rate": 0.0024794520547945205, | |
| "loss": 0.3491, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 5.1525423728813555, | |
| "grad_norm": 2.7462785243988037, | |
| "learning_rate": 0.002410958904109589, | |
| "loss": 0.4279, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 5.288135593220339, | |
| "grad_norm": 0.1708032637834549, | |
| "learning_rate": 0.0023424657534246575, | |
| "loss": 0.3868, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 5.423728813559322, | |
| "grad_norm": 0.4240102469921112, | |
| "learning_rate": 0.002273972602739726, | |
| "loss": 0.416, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 5.559322033898305, | |
| "grad_norm": 0.3024919331073761, | |
| "learning_rate": 0.002205479452054795, | |
| "loss": 0.3745, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 5.694915254237288, | |
| "grad_norm": 0.4930131733417511, | |
| "learning_rate": 0.002136986301369863, | |
| "loss": 0.4183, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 5.830508474576272, | |
| "grad_norm": 2.3017737865448, | |
| "learning_rate": 0.002068493150684932, | |
| "loss": 0.3791, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 5.966101694915254, | |
| "grad_norm": 1.3229153156280518, | |
| "learning_rate": 0.002, | |
| "loss": 0.3887, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 5.9932203389830505, | |
| "eval_accuracy": 0.8645038167938931, | |
| "eval_f1": 0.8382821499206712, | |
| "eval_loss": 0.36667701601982117, | |
| "eval_precision": 0.8196564265766058, | |
| "eval_recall": 0.8748833666571456, | |
| "eval_runtime": 3.2223, | |
| "eval_samples_per_second": 162.616, | |
| "eval_steps_per_second": 10.241, | |
| "step": 442 | |
| }, | |
| { | |
| "epoch": 6.101694915254237, | |
| "grad_norm": 1.612464189529419, | |
| "learning_rate": 0.0019315068493150687, | |
| "loss": 0.3626, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 6.237288135593221, | |
| "grad_norm": 0.30710288882255554, | |
| "learning_rate": 0.0018630136986301372, | |
| "loss": 0.4014, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 6.372881355932203, | |
| "grad_norm": 1.5897902250289917, | |
| "learning_rate": 0.0017945205479452057, | |
| "loss": 0.3953, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 6.508474576271187, | |
| "grad_norm": 1.6094199419021606, | |
| "learning_rate": 0.0017260273972602741, | |
| "loss": 0.4189, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 6.6440677966101696, | |
| "grad_norm": 1.078356385231018, | |
| "learning_rate": 0.0016575342465753426, | |
| "loss": 0.4268, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 6.779661016949152, | |
| "grad_norm": 2.4723339080810547, | |
| "learning_rate": 0.001589041095890411, | |
| "loss": 0.4025, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 6.915254237288136, | |
| "grad_norm": 1.7719030380249023, | |
| "learning_rate": 0.0015205479452054796, | |
| "loss": 0.3947, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 6.996610169491525, | |
| "eval_accuracy": 0.8530534351145038, | |
| "eval_f1": 0.8282881157569955, | |
| "eval_loss": 0.39514508843421936, | |
| "eval_precision": 0.8098487236054208, | |
| "eval_recall": 0.8744263543749404, | |
| "eval_runtime": 2.9797, | |
| "eval_samples_per_second": 175.856, | |
| "eval_steps_per_second": 11.075, | |
| "step": 516 | |
| }, | |
| { | |
| "epoch": 7.0508474576271185, | |
| "grad_norm": 0.9901585578918457, | |
| "learning_rate": 0.0014520547945205478, | |
| "loss": 0.364, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 7.186440677966102, | |
| "grad_norm": 1.4664475917816162, | |
| "learning_rate": 0.0013835616438356163, | |
| "loss": 0.3738, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 7.322033898305085, | |
| "grad_norm": 1.8693777322769165, | |
| "learning_rate": 0.0013150684931506848, | |
| "loss": 0.409, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 7.4576271186440675, | |
| "grad_norm": 1.6105625629425049, | |
| "learning_rate": 0.0012465753424657534, | |
| "loss": 0.373, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 7.593220338983051, | |
| "grad_norm": 1.700050950050354, | |
| "learning_rate": 0.001178082191780822, | |
| "loss": 0.3899, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 7.728813559322034, | |
| "grad_norm": 0.6655128598213196, | |
| "learning_rate": 0.0011095890410958904, | |
| "loss": 0.3573, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 7.864406779661017, | |
| "grad_norm": 2.1642274856567383, | |
| "learning_rate": 0.0010410958904109589, | |
| "loss": 0.3828, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 1.8888564109802246, | |
| "learning_rate": 0.0009726027397260274, | |
| "loss": 0.3741, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8683206106870229, | |
| "eval_f1": 0.8398298890291713, | |
| "eval_loss": 0.34485381841659546, | |
| "eval_precision": 0.8229333868378812, | |
| "eval_recall": 0.8677806341045415, | |
| "eval_runtime": 2.988, | |
| "eval_samples_per_second": 175.367, | |
| "eval_steps_per_second": 11.044, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 8.135593220338983, | |
| "grad_norm": 1.7316805124282837, | |
| "learning_rate": 0.0009041095890410959, | |
| "loss": 0.4155, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 8.271186440677965, | |
| "grad_norm": 0.8091392517089844, | |
| "learning_rate": 0.0008356164383561645, | |
| "loss": 0.3834, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 8.40677966101695, | |
| "grad_norm": 0.47697487473487854, | |
| "learning_rate": 0.000767123287671233, | |
| "loss": 0.3501, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 8.542372881355933, | |
| "grad_norm": 0.419321209192276, | |
| "learning_rate": 0.0006986301369863014, | |
| "loss": 0.349, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 8.677966101694915, | |
| "grad_norm": 2.6489877700805664, | |
| "learning_rate": 0.0006301369863013699, | |
| "loss": 0.3791, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 8.813559322033898, | |
| "grad_norm": 0.7624053359031677, | |
| "learning_rate": 0.0005616438356164384, | |
| "loss": 0.3655, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 8.94915254237288, | |
| "grad_norm": 0.44931384921073914, | |
| "learning_rate": 0.0004931506849315068, | |
| "loss": 0.3964, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 8.989830508474576, | |
| "eval_accuracy": 0.8587786259541985, | |
| "eval_f1": 0.830509659935309, | |
| "eval_loss": 0.3624710738658905, | |
| "eval_precision": 0.8127875562832583, | |
| "eval_recall": 0.8637722555460345, | |
| "eval_runtime": 3.2987, | |
| "eval_samples_per_second": 158.852, | |
| "eval_steps_per_second": 10.004, | |
| "step": 663 | |
| }, | |
| { | |
| "epoch": 9.084745762711865, | |
| "grad_norm": 0.7742342948913574, | |
| "learning_rate": 0.00042465753424657536, | |
| "loss": 0.3609, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 9.220338983050848, | |
| "grad_norm": 0.2361624389886856, | |
| "learning_rate": 0.00035616438356164383, | |
| "loss": 0.3231, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 9.35593220338983, | |
| "grad_norm": 0.3743881583213806, | |
| "learning_rate": 0.00028767123287671236, | |
| "loss": 0.3835, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 9.491525423728813, | |
| "grad_norm": 0.3935371935367584, | |
| "learning_rate": 0.00021917808219178083, | |
| "loss": 0.3503, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 9.627118644067796, | |
| "grad_norm": 0.6624675393104553, | |
| "learning_rate": 0.00015068493150684933, | |
| "loss": 0.3023, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 9.76271186440678, | |
| "grad_norm": 1.4423840045928955, | |
| "learning_rate": 8.21917808219178e-05, | |
| "loss": 0.404, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 9.898305084745763, | |
| "grad_norm": 0.639860212802887, | |
| "learning_rate": 1.3698630136986302e-05, | |
| "loss": 0.3845, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 9.898305084745763, | |
| "eval_accuracy": 0.8568702290076335, | |
| "eval_f1": 0.8291712851274695, | |
| "eval_loss": 0.3568853735923767, | |
| "eval_precision": 0.8111482756381113, | |
| "eval_recall": 0.8649052651623346, | |
| "eval_runtime": 3.0816, | |
| "eval_samples_per_second": 170.043, | |
| "eval_steps_per_second": 10.709, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 9.898305084745763, | |
| "step": 730, | |
| "total_flos": 3.6354082184735293e+18, | |
| "train_loss": 0.42946596570210915, | |
| "train_runtime": 617.6458, | |
| "train_samples_per_second": 76.225, | |
| "train_steps_per_second": 1.182 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 730, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "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": 3.6354082184735293e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |