| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.7284837131855552, | |
| "eval_steps": 100, | |
| "global_step": 7000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.010406910188365074, | |
| "grad_norm": 17.59031867980957, | |
| "learning_rate": 4.982655149686058e-05, | |
| "loss": 4.0667, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.010406910188365074, | |
| "eval_loss": 3.950298309326172, | |
| "eval_runtime": 3517.0273, | |
| "eval_samples_per_second": 11.923, | |
| "eval_steps_per_second": 1.49, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.020813820376730147, | |
| "grad_norm": 12.363835334777832, | |
| "learning_rate": 4.965310299372117e-05, | |
| "loss": 3.6567, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.020813820376730147, | |
| "eval_loss": 3.9310147762298584, | |
| "eval_runtime": 3451.7016, | |
| "eval_samples_per_second": 12.148, | |
| "eval_steps_per_second": 1.519, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.031220730565095223, | |
| "grad_norm": 9.739866256713867, | |
| "learning_rate": 4.947965449058175e-05, | |
| "loss": 3.6612, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.031220730565095223, | |
| "eval_loss": 3.9458096027374268, | |
| "eval_runtime": 3450.7573, | |
| "eval_samples_per_second": 12.152, | |
| "eval_steps_per_second": 1.519, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.041627640753460295, | |
| "grad_norm": 7.430461406707764, | |
| "learning_rate": 4.930620598744233e-05, | |
| "loss": 3.6409, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.041627640753460295, | |
| "eval_loss": 3.928576707839966, | |
| "eval_runtime": 3466.9688, | |
| "eval_samples_per_second": 12.095, | |
| "eval_steps_per_second": 1.512, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.052034550941825373, | |
| "grad_norm": 9.64016056060791, | |
| "learning_rate": 4.913275748430291e-05, | |
| "loss": 3.5862, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.052034550941825373, | |
| "eval_loss": 3.9448041915893555, | |
| "eval_runtime": 3452.0797, | |
| "eval_samples_per_second": 12.147, | |
| "eval_steps_per_second": 1.519, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.062441461130190445, | |
| "grad_norm": 8.797385215759277, | |
| "learning_rate": 4.8959308981163494e-05, | |
| "loss": 3.5627, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.062441461130190445, | |
| "eval_loss": 3.925294876098633, | |
| "eval_runtime": 3449.9288, | |
| "eval_samples_per_second": 12.154, | |
| "eval_steps_per_second": 1.519, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.07284837131855552, | |
| "grad_norm": 7.623824596405029, | |
| "learning_rate": 4.8785860478024074e-05, | |
| "loss": 3.5507, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.07284837131855552, | |
| "eval_loss": 3.9132120609283447, | |
| "eval_runtime": 3475.0918, | |
| "eval_samples_per_second": 12.066, | |
| "eval_steps_per_second": 1.508, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.08325528150692059, | |
| "grad_norm": 7.819701671600342, | |
| "learning_rate": 4.861241197488466e-05, | |
| "loss": 3.4453, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.08325528150692059, | |
| "eval_loss": 3.9031147956848145, | |
| "eval_runtime": 3466.7274, | |
| "eval_samples_per_second": 12.096, | |
| "eval_steps_per_second": 1.512, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.09366219169528567, | |
| "grad_norm": 7.915994167327881, | |
| "learning_rate": 4.843896347174524e-05, | |
| "loss": 3.4908, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.09366219169528567, | |
| "eval_loss": 3.922236442565918, | |
| "eval_runtime": 3474.1465, | |
| "eval_samples_per_second": 12.07, | |
| "eval_steps_per_second": 1.509, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.10406910188365075, | |
| "grad_norm": 6.398956775665283, | |
| "learning_rate": 4.826551496860582e-05, | |
| "loss": 3.4467, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.10406910188365075, | |
| "eval_loss": 3.9016082286834717, | |
| "eval_runtime": 3439.9815, | |
| "eval_samples_per_second": 12.19, | |
| "eval_steps_per_second": 1.524, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.11447601207201581, | |
| "grad_norm": 5.43521785736084, | |
| "learning_rate": 4.8092066465466404e-05, | |
| "loss": 3.4598, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.11447601207201581, | |
| "eval_loss": 3.9063823223114014, | |
| "eval_runtime": 3446.7857, | |
| "eval_samples_per_second": 12.166, | |
| "eval_steps_per_second": 1.521, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.12488292226038089, | |
| "grad_norm": 5.717388153076172, | |
| "learning_rate": 4.791861796232699e-05, | |
| "loss": 3.4359, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.12488292226038089, | |
| "eval_loss": 3.881629705429077, | |
| "eval_runtime": 3449.577, | |
| "eval_samples_per_second": 12.156, | |
| "eval_steps_per_second": 1.52, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.13528983244874596, | |
| "grad_norm": 5.882245063781738, | |
| "learning_rate": 4.774516945918757e-05, | |
| "loss": 3.425, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.13528983244874596, | |
| "eval_loss": 3.883547782897949, | |
| "eval_runtime": 3446.1151, | |
| "eval_samples_per_second": 12.168, | |
| "eval_steps_per_second": 1.521, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.14569674263711105, | |
| "grad_norm": 7.307028293609619, | |
| "learning_rate": 4.7571720956048146e-05, | |
| "loss": 3.4218, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.14569674263711105, | |
| "eval_loss": 3.8712680339813232, | |
| "eval_runtime": 3438.5402, | |
| "eval_samples_per_second": 12.195, | |
| "eval_steps_per_second": 1.524, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.1561036528254761, | |
| "grad_norm": 7.0670485496521, | |
| "learning_rate": 4.7398272452908734e-05, | |
| "loss": 3.3977, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.1561036528254761, | |
| "eval_loss": 3.8845202922821045, | |
| "eval_runtime": 3441.119, | |
| "eval_samples_per_second": 12.186, | |
| "eval_steps_per_second": 1.523, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.16651056301384118, | |
| "grad_norm": 5.122912883758545, | |
| "learning_rate": 4.7224823949769314e-05, | |
| "loss": 3.4539, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.16651056301384118, | |
| "eval_loss": 3.8676364421844482, | |
| "eval_runtime": 3435.2798, | |
| "eval_samples_per_second": 12.206, | |
| "eval_steps_per_second": 1.526, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.17691747320220627, | |
| "grad_norm": 8.177063941955566, | |
| "learning_rate": 4.7051375446629895e-05, | |
| "loss": 3.4278, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.17691747320220627, | |
| "eval_loss": 3.8678488731384277, | |
| "eval_runtime": 3414.3552, | |
| "eval_samples_per_second": 12.281, | |
| "eval_steps_per_second": 1.535, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.18732438339057134, | |
| "grad_norm": 5.061961650848389, | |
| "learning_rate": 4.687792694349048e-05, | |
| "loss": 3.4475, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.18732438339057134, | |
| "eval_loss": 3.8584587574005127, | |
| "eval_runtime": 3487.8876, | |
| "eval_samples_per_second": 12.022, | |
| "eval_steps_per_second": 1.503, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.1977312935789364, | |
| "grad_norm": 5.108020305633545, | |
| "learning_rate": 4.670447844035106e-05, | |
| "loss": 3.4618, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.1977312935789364, | |
| "eval_loss": 3.869356155395508, | |
| "eval_runtime": 3642.3755, | |
| "eval_samples_per_second": 11.512, | |
| "eval_steps_per_second": 1.439, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.2081382037673015, | |
| "grad_norm": 5.9767351150512695, | |
| "learning_rate": 4.6531029937211644e-05, | |
| "loss": 3.3986, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.2081382037673015, | |
| "eval_loss": 3.8760101795196533, | |
| "eval_runtime": 3500.4112, | |
| "eval_samples_per_second": 11.979, | |
| "eval_steps_per_second": 1.498, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.21854511395566656, | |
| "grad_norm": 7.525568008422852, | |
| "learning_rate": 4.6357581434072225e-05, | |
| "loss": 3.3452, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.21854511395566656, | |
| "eval_loss": 3.8635501861572266, | |
| "eval_runtime": 3434.6174, | |
| "eval_samples_per_second": 12.209, | |
| "eval_steps_per_second": 1.526, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.22895202414403162, | |
| "grad_norm": 6.358871936798096, | |
| "learning_rate": 4.618413293093281e-05, | |
| "loss": 3.3387, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.22895202414403162, | |
| "eval_loss": 3.85604190826416, | |
| "eval_runtime": 3437.498, | |
| "eval_samples_per_second": 12.198, | |
| "eval_steps_per_second": 1.525, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.23935893433239672, | |
| "grad_norm": 7.322586536407471, | |
| "learning_rate": 4.6010684427793386e-05, | |
| "loss": 3.3527, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.23935893433239672, | |
| "eval_loss": 3.867178440093994, | |
| "eval_runtime": 3435.5675, | |
| "eval_samples_per_second": 12.205, | |
| "eval_steps_per_second": 1.526, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.24976584452076178, | |
| "grad_norm": 8.45722770690918, | |
| "learning_rate": 4.5837235924653974e-05, | |
| "loss": 3.4033, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.24976584452076178, | |
| "eval_loss": 3.8560798168182373, | |
| "eval_runtime": 3565.6947, | |
| "eval_samples_per_second": 11.76, | |
| "eval_steps_per_second": 1.47, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.2601727547091269, | |
| "grad_norm": 5.179561614990234, | |
| "learning_rate": 4.5663787421514554e-05, | |
| "loss": 3.4205, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.2601727547091269, | |
| "eval_loss": 3.8475325107574463, | |
| "eval_runtime": 3647.5472, | |
| "eval_samples_per_second": 11.496, | |
| "eval_steps_per_second": 1.437, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.2705796648974919, | |
| "grad_norm": 5.7942399978637695, | |
| "learning_rate": 4.5490338918375135e-05, | |
| "loss": 3.2885, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.2705796648974919, | |
| "eval_loss": 3.8502981662750244, | |
| "eval_runtime": 3640.4425, | |
| "eval_samples_per_second": 11.518, | |
| "eval_steps_per_second": 1.44, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.280986575085857, | |
| "grad_norm": 6.213371753692627, | |
| "learning_rate": 4.5316890415235716e-05, | |
| "loss": 3.3147, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.280986575085857, | |
| "eval_loss": 3.856261730194092, | |
| "eval_runtime": 3643.059, | |
| "eval_samples_per_second": 11.51, | |
| "eval_steps_per_second": 1.439, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.2913934852742221, | |
| "grad_norm": 4.848277568817139, | |
| "learning_rate": 4.5143441912096303e-05, | |
| "loss": 3.3275, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.2913934852742221, | |
| "eval_loss": 3.8547070026397705, | |
| "eval_runtime": 3641.446, | |
| "eval_samples_per_second": 11.515, | |
| "eval_steps_per_second": 1.44, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.30180039546258713, | |
| "grad_norm": 6.599040508270264, | |
| "learning_rate": 4.4969993408956884e-05, | |
| "loss": 3.3368, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.30180039546258713, | |
| "eval_loss": 3.8605613708496094, | |
| "eval_runtime": 3525.3985, | |
| "eval_samples_per_second": 11.894, | |
| "eval_steps_per_second": 1.487, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.3122073056509522, | |
| "grad_norm": 7.062480449676514, | |
| "learning_rate": 4.4796544905817465e-05, | |
| "loss": 3.3218, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.3122073056509522, | |
| "eval_loss": 3.8569867610931396, | |
| "eval_runtime": 3438.6282, | |
| "eval_samples_per_second": 12.194, | |
| "eval_steps_per_second": 1.524, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.3226142158393173, | |
| "grad_norm": 5.044654846191406, | |
| "learning_rate": 4.462309640267805e-05, | |
| "loss": 3.2919, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 0.3226142158393173, | |
| "eval_loss": 3.857635974884033, | |
| "eval_runtime": 3422.615, | |
| "eval_samples_per_second": 12.251, | |
| "eval_steps_per_second": 1.532, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 0.33302112602768236, | |
| "grad_norm": 4.636962413787842, | |
| "learning_rate": 4.4449647899538626e-05, | |
| "loss": 3.2528, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.33302112602768236, | |
| "eval_loss": 3.860072612762451, | |
| "eval_runtime": 3427.8634, | |
| "eval_samples_per_second": 12.233, | |
| "eval_steps_per_second": 1.529, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.34342803621604745, | |
| "grad_norm": 6.344686985015869, | |
| "learning_rate": 4.427619939639921e-05, | |
| "loss": 3.3591, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 0.34342803621604745, | |
| "eval_loss": 3.8569118976593018, | |
| "eval_runtime": 3432.7441, | |
| "eval_samples_per_second": 12.215, | |
| "eval_steps_per_second": 1.527, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 0.35383494640441254, | |
| "grad_norm": 5.941151142120361, | |
| "learning_rate": 4.4102750893259795e-05, | |
| "loss": 3.3384, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 0.35383494640441254, | |
| "eval_loss": 3.851085901260376, | |
| "eval_runtime": 3432.068, | |
| "eval_samples_per_second": 12.218, | |
| "eval_steps_per_second": 1.527, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 0.3642418565927776, | |
| "grad_norm": 6.193995475769043, | |
| "learning_rate": 4.3929302390120375e-05, | |
| "loss": 3.2801, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.3642418565927776, | |
| "eval_loss": 3.843860387802124, | |
| "eval_runtime": 3429.2875, | |
| "eval_samples_per_second": 12.228, | |
| "eval_steps_per_second": 1.529, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.3746487667811427, | |
| "grad_norm": 7.5851545333862305, | |
| "learning_rate": 4.3755853886980956e-05, | |
| "loss": 3.2912, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 0.3746487667811427, | |
| "eval_loss": 3.851679563522339, | |
| "eval_runtime": 3429.968, | |
| "eval_samples_per_second": 12.225, | |
| "eval_steps_per_second": 1.528, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 0.38505567696950777, | |
| "grad_norm": 5.366103172302246, | |
| "learning_rate": 4.3582405383841544e-05, | |
| "loss": 3.2847, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 0.38505567696950777, | |
| "eval_loss": 3.847703218460083, | |
| "eval_runtime": 3435.5138, | |
| "eval_samples_per_second": 12.205, | |
| "eval_steps_per_second": 1.526, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 0.3954625871578728, | |
| "grad_norm": 5.834432125091553, | |
| "learning_rate": 4.3408956880702124e-05, | |
| "loss": 3.3753, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 0.3954625871578728, | |
| "eval_loss": 3.8320815563201904, | |
| "eval_runtime": 3440.1983, | |
| "eval_samples_per_second": 12.189, | |
| "eval_steps_per_second": 1.524, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 0.4058694973462379, | |
| "grad_norm": 4.3302178382873535, | |
| "learning_rate": 4.3235508377562705e-05, | |
| "loss": 3.2613, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 0.4058694973462379, | |
| "eval_loss": 3.8347744941711426, | |
| "eval_runtime": 3475.7623, | |
| "eval_samples_per_second": 12.064, | |
| "eval_steps_per_second": 1.508, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 0.416276407534603, | |
| "grad_norm": 10.435820579528809, | |
| "learning_rate": 4.3062059874423286e-05, | |
| "loss": 3.2497, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.416276407534603, | |
| "eval_loss": 3.835855484008789, | |
| "eval_runtime": 3464.7097, | |
| "eval_samples_per_second": 12.103, | |
| "eval_steps_per_second": 1.513, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.426683317722968, | |
| "grad_norm": 6.157552242279053, | |
| "learning_rate": 4.2888611371283866e-05, | |
| "loss": 3.3412, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 0.426683317722968, | |
| "eval_loss": 3.834092140197754, | |
| "eval_runtime": 3534.2463, | |
| "eval_samples_per_second": 11.864, | |
| "eval_steps_per_second": 1.483, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 0.4370902279113331, | |
| "grad_norm": 6.643843650817871, | |
| "learning_rate": 4.271516286814445e-05, | |
| "loss": 3.2945, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 0.4370902279113331, | |
| "eval_loss": 3.8426012992858887, | |
| "eval_runtime": 4149.9023, | |
| "eval_samples_per_second": 10.104, | |
| "eval_steps_per_second": 1.263, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 0.4474971380996982, | |
| "grad_norm": 6.451257705688477, | |
| "learning_rate": 4.254171436500503e-05, | |
| "loss": 3.2203, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 0.4474971380996982, | |
| "eval_loss": 3.840031147003174, | |
| "eval_runtime": 4060.1081, | |
| "eval_samples_per_second": 10.328, | |
| "eval_steps_per_second": 1.291, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 0.45790404828806325, | |
| "grad_norm": 7.092510223388672, | |
| "learning_rate": 4.2368265861865615e-05, | |
| "loss": 3.2779, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 0.45790404828806325, | |
| "eval_loss": 3.8369994163513184, | |
| "eval_runtime": 4092.0714, | |
| "eval_samples_per_second": 10.247, | |
| "eval_steps_per_second": 1.281, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 0.46831095847642834, | |
| "grad_norm": 6.0258588790893555, | |
| "learning_rate": 4.2194817358726196e-05, | |
| "loss": 3.1867, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.46831095847642834, | |
| "eval_loss": 3.837413787841797, | |
| "eval_runtime": 4068.8274, | |
| "eval_samples_per_second": 10.306, | |
| "eval_steps_per_second": 1.288, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.47871786866479343, | |
| "grad_norm": 6.42568302154541, | |
| "learning_rate": 4.202136885558678e-05, | |
| "loss": 3.2981, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 0.47871786866479343, | |
| "eval_loss": 3.8490710258483887, | |
| "eval_runtime": 4072.501, | |
| "eval_samples_per_second": 10.296, | |
| "eval_steps_per_second": 1.287, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 0.48912477885315847, | |
| "grad_norm": 4.5895466804504395, | |
| "learning_rate": 4.1847920352447364e-05, | |
| "loss": 3.2523, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 0.48912477885315847, | |
| "eval_loss": 3.8223540782928467, | |
| "eval_runtime": 3505.4411, | |
| "eval_samples_per_second": 11.962, | |
| "eval_steps_per_second": 1.495, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 0.49953168904152356, | |
| "grad_norm": 4.485264778137207, | |
| "learning_rate": 4.1674471849307945e-05, | |
| "loss": 3.2375, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 0.49953168904152356, | |
| "eval_loss": 3.847806692123413, | |
| "eval_runtime": 3468.0824, | |
| "eval_samples_per_second": 12.091, | |
| "eval_steps_per_second": 1.511, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 0.5099385992298886, | |
| "grad_norm": 6.584381103515625, | |
| "learning_rate": 4.1501023346168526e-05, | |
| "loss": 3.2327, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 0.5099385992298886, | |
| "eval_loss": 3.849238872528076, | |
| "eval_runtime": 3470.3976, | |
| "eval_samples_per_second": 12.083, | |
| "eval_steps_per_second": 1.51, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 0.5203455094182537, | |
| "grad_norm": 7.7780303955078125, | |
| "learning_rate": 4.1327574843029107e-05, | |
| "loss": 3.2209, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.5203455094182537, | |
| "eval_loss": 3.855090856552124, | |
| "eval_runtime": 3470.2986, | |
| "eval_samples_per_second": 12.083, | |
| "eval_steps_per_second": 1.511, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.5307524196066188, | |
| "grad_norm": 7.771067142486572, | |
| "learning_rate": 4.115412633988969e-05, | |
| "loss": 3.1719, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 0.5307524196066188, | |
| "eval_loss": 3.8357627391815186, | |
| "eval_runtime": 3461.6324, | |
| "eval_samples_per_second": 12.113, | |
| "eval_steps_per_second": 1.514, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 0.5411593297949838, | |
| "grad_norm": 4.688082218170166, | |
| "learning_rate": 4.098067783675027e-05, | |
| "loss": 3.2683, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 0.5411593297949838, | |
| "eval_loss": 3.847905158996582, | |
| "eval_runtime": 3464.2184, | |
| "eval_samples_per_second": 12.104, | |
| "eval_steps_per_second": 1.513, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 0.551566239983349, | |
| "grad_norm": 5.934951305389404, | |
| "learning_rate": 4.0807229333610856e-05, | |
| "loss": 3.2195, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 0.551566239983349, | |
| "eval_loss": 3.8351073265075684, | |
| "eval_runtime": 3462.7613, | |
| "eval_samples_per_second": 12.109, | |
| "eval_steps_per_second": 1.514, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 0.561973150171714, | |
| "grad_norm": 8.332444190979004, | |
| "learning_rate": 4.0633780830471436e-05, | |
| "loss": 3.2279, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 0.561973150171714, | |
| "eval_loss": 3.827998638153076, | |
| "eval_runtime": 3460.3148, | |
| "eval_samples_per_second": 12.118, | |
| "eval_steps_per_second": 1.515, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 0.572380060360079, | |
| "grad_norm": 4.6214599609375, | |
| "learning_rate": 4.046033232733202e-05, | |
| "loss": 3.2472, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.572380060360079, | |
| "eval_loss": 3.8354616165161133, | |
| "eval_runtime": 3550.6038, | |
| "eval_samples_per_second": 11.81, | |
| "eval_steps_per_second": 1.476, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.5827869705484442, | |
| "grad_norm": 7.2509589195251465, | |
| "learning_rate": 4.02868838241926e-05, | |
| "loss": 3.2335, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 0.5827869705484442, | |
| "eval_loss": 3.829226016998291, | |
| "eval_runtime": 3930.9878, | |
| "eval_samples_per_second": 10.667, | |
| "eval_steps_per_second": 1.334, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 0.5931938807368092, | |
| "grad_norm": 5.364116191864014, | |
| "learning_rate": 4.0113435321053185e-05, | |
| "loss": 3.1955, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 0.5931938807368092, | |
| "eval_loss": 3.8261866569519043, | |
| "eval_runtime": 3914.6993, | |
| "eval_samples_per_second": 10.711, | |
| "eval_steps_per_second": 1.339, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 0.6036007909251743, | |
| "grad_norm": 7.65122652053833, | |
| "learning_rate": 3.9939986817913766e-05, | |
| "loss": 3.1441, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 0.6036007909251743, | |
| "eval_loss": 3.842827796936035, | |
| "eval_runtime": 3638.8184, | |
| "eval_samples_per_second": 11.524, | |
| "eval_steps_per_second": 1.441, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 0.6140077011135394, | |
| "grad_norm": 5.782172203063965, | |
| "learning_rate": 3.976653831477434e-05, | |
| "loss": 3.1673, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 0.6140077011135394, | |
| "eval_loss": 3.8224923610687256, | |
| "eval_runtime": 3852.8778, | |
| "eval_samples_per_second": 10.883, | |
| "eval_steps_per_second": 1.361, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 0.6244146113019045, | |
| "grad_norm": 5.476900100708008, | |
| "learning_rate": 3.959308981163493e-05, | |
| "loss": 3.12, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.6244146113019045, | |
| "eval_loss": 3.8422653675079346, | |
| "eval_runtime": 3460.5604, | |
| "eval_samples_per_second": 12.117, | |
| "eval_steps_per_second": 1.515, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.6348215214902695, | |
| "grad_norm": 4.047117710113525, | |
| "learning_rate": 3.941964130849551e-05, | |
| "loss": 3.1838, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 0.6348215214902695, | |
| "eval_loss": 3.8332009315490723, | |
| "eval_runtime": 3464.3723, | |
| "eval_samples_per_second": 12.104, | |
| "eval_steps_per_second": 1.513, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 0.6452284316786346, | |
| "grad_norm": 5.556035995483398, | |
| "learning_rate": 3.924619280535609e-05, | |
| "loss": 3.1586, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 0.6452284316786346, | |
| "eval_loss": 3.8465607166290283, | |
| "eval_runtime": 3461.4127, | |
| "eval_samples_per_second": 12.114, | |
| "eval_steps_per_second": 1.514, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 0.6556353418669997, | |
| "grad_norm": 5.888989448547363, | |
| "learning_rate": 3.9072744302216676e-05, | |
| "loss": 3.1398, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 0.6556353418669997, | |
| "eval_loss": 3.8334028720855713, | |
| "eval_runtime": 3490.2152, | |
| "eval_samples_per_second": 12.014, | |
| "eval_steps_per_second": 1.502, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 0.6660422520553647, | |
| "grad_norm": 6.468564987182617, | |
| "learning_rate": 3.889929579907726e-05, | |
| "loss": 3.193, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 0.6660422520553647, | |
| "eval_loss": 3.853891372680664, | |
| "eval_runtime": 3456.8398, | |
| "eval_samples_per_second": 12.13, | |
| "eval_steps_per_second": 1.516, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 0.6764491622437299, | |
| "grad_norm": 5.189215660095215, | |
| "learning_rate": 3.872584729593784e-05, | |
| "loss": 3.204, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.6764491622437299, | |
| "eval_loss": 3.8300530910491943, | |
| "eval_runtime": 3457.5384, | |
| "eval_samples_per_second": 12.128, | |
| "eval_steps_per_second": 1.516, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.6868560724320949, | |
| "grad_norm": 5.385128021240234, | |
| "learning_rate": 3.8552398792798425e-05, | |
| "loss": 3.0612, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 0.6868560724320949, | |
| "eval_loss": 3.8288888931274414, | |
| "eval_runtime": 3453.9786, | |
| "eval_samples_per_second": 12.14, | |
| "eval_steps_per_second": 1.518, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 0.6972629826204599, | |
| "grad_norm": 4.424393653869629, | |
| "learning_rate": 3.8378950289659006e-05, | |
| "loss": 3.1726, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 0.6972629826204599, | |
| "eval_loss": 3.820075511932373, | |
| "eval_runtime": 3454.0788, | |
| "eval_samples_per_second": 12.14, | |
| "eval_steps_per_second": 1.518, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 0.7076698928088251, | |
| "grad_norm": 3.365438222885132, | |
| "learning_rate": 3.820550178651958e-05, | |
| "loss": 3.1238, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 0.7076698928088251, | |
| "eval_loss": 3.8395471572875977, | |
| "eval_runtime": 3808.8554, | |
| "eval_samples_per_second": 11.009, | |
| "eval_steps_per_second": 1.376, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 0.7180768029971901, | |
| "grad_norm": 5.473466873168945, | |
| "learning_rate": 3.803205328338017e-05, | |
| "loss": 3.198, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 0.7180768029971901, | |
| "eval_loss": 3.831690549850464, | |
| "eval_runtime": 4160.6602, | |
| "eval_samples_per_second": 10.078, | |
| "eval_steps_per_second": 1.26, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 0.7284837131855552, | |
| "grad_norm": 4.136599540710449, | |
| "learning_rate": 3.785860478024075e-05, | |
| "loss": 3.1025, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.7284837131855552, | |
| "eval_loss": 3.831153631210327, | |
| "eval_runtime": 3452.4147, | |
| "eval_samples_per_second": 12.146, | |
| "eval_steps_per_second": 1.518, | |
| "step": 7000 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 28827, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 1000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.4632353792e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |