| { |
| "best_global_step": 600, |
| "best_metric": 0.33704105019569397, |
| "best_model_checkpoint": "/workspace/anchor/qwen-compassion-s2/train/checkpoint-600", |
| "epoch": 1.0, |
| "eval_steps": 50, |
| "global_step": 614, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.008143322475570033, |
| "grad_norm": 47.063079833984375, |
| "learning_rate": 2.5806451612903226e-05, |
| "loss": 3.9245, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.016286644951140065, |
| "grad_norm": 26.774829864501953, |
| "learning_rate": 5.8064516129032266e-05, |
| "loss": 3.1178, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.024429967426710098, |
| "grad_norm": 6.747068405151367, |
| "learning_rate": 9.032258064516129e-05, |
| "loss": 2.3953, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.03257328990228013, |
| "grad_norm": 3.2441258430480957, |
| "learning_rate": 0.00012258064516129034, |
| "loss": 1.5483, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.04071661237785016, |
| "grad_norm": 3.071211814880371, |
| "learning_rate": 0.00015483870967741937, |
| "loss": 1.0951, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.048859934853420196, |
| "grad_norm": 6.013782501220703, |
| "learning_rate": 0.0001870967741935484, |
| "loss": 0.8924, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.057003257328990226, |
| "grad_norm": 4.459931373596191, |
| "learning_rate": 0.00019998693330688282, |
| "loss": 0.7041, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.06514657980456026, |
| "grad_norm": 4.017563343048096, |
| "learning_rate": 0.00019990709365929677, |
| "loss": 0.629, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.0732899022801303, |
| "grad_norm": 2.9156265258789062, |
| "learning_rate": 0.00019975473152335726, |
| "loss": 0.6571, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.08143322475570032, |
| "grad_norm": 2.0442211627960205, |
| "learning_rate": 0.0001995299574986306, |
| "loss": 0.6642, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.08143322475570032, |
| "eval_loss": 0.5125934481620789, |
| "eval_runtime": 165.6729, |
| "eval_samples_per_second": 1.521, |
| "eval_steps_per_second": 0.761, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.08957654723127036, |
| "grad_norm": 3.1159253120422363, |
| "learning_rate": 0.00019923293474842174, |
| "loss": 0.5377, |
| "step": 55 |
| }, |
| { |
| "epoch": 0.09771986970684039, |
| "grad_norm": 1.3203703165054321, |
| "learning_rate": 0.00019886387888133413, |
| "loss": 0.5609, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.10586319218241043, |
| "grad_norm": 1.0699498653411865, |
| "learning_rate": 0.00019842305779475968, |
| "loss": 0.5683, |
| "step": 65 |
| }, |
| { |
| "epoch": 0.11400651465798045, |
| "grad_norm": 1.2404087781906128, |
| "learning_rate": 0.0001979107914804122, |
| "loss": 0.5308, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.12214983713355049, |
| "grad_norm": 1.3837926387786865, |
| "learning_rate": 0.00019732745179204552, |
| "loss": 0.5921, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.13029315960912052, |
| "grad_norm": 1.1491721868515015, |
| "learning_rate": 0.00019667346217552527, |
| "loss": 0.4811, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.13843648208469056, |
| "grad_norm": 1.709602952003479, |
| "learning_rate": 0.00019594929736144976, |
| "loss": 0.5148, |
| "step": 85 |
| }, |
| { |
| "epoch": 0.1465798045602606, |
| "grad_norm": 1.2207844257354736, |
| "learning_rate": 0.00019515548302054335, |
| "loss": 0.6146, |
| "step": 90 |
| }, |
| { |
| "epoch": 0.15472312703583063, |
| "grad_norm": 1.2277162075042725, |
| "learning_rate": 0.0001942925953820725, |
| "loss": 0.5356, |
| "step": 95 |
| }, |
| { |
| "epoch": 0.16286644951140064, |
| "grad_norm": 1.0094375610351562, |
| "learning_rate": 0.00019336126081556134, |
| "loss": 0.5516, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16286644951140064, |
| "eval_loss": 0.4084537923336029, |
| "eval_runtime": 84.8676, |
| "eval_samples_per_second": 2.969, |
| "eval_steps_per_second": 1.485, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.17100977198697068, |
| "grad_norm": 1.2880449295043945, |
| "learning_rate": 0.00019236215537611046, |
| "loss": 0.4939, |
| "step": 105 |
| }, |
| { |
| "epoch": 0.1791530944625407, |
| "grad_norm": 1.0443342924118042, |
| "learning_rate": 0.00019129600431364897, |
| "loss": 0.4457, |
| "step": 110 |
| }, |
| { |
| "epoch": 0.18729641693811075, |
| "grad_norm": 1.0998979806900024, |
| "learning_rate": 0.00019016358154647618, |
| "loss": 0.5006, |
| "step": 115 |
| }, |
| { |
| "epoch": 0.19543973941368079, |
| "grad_norm": 1.535317301750183, |
| "learning_rate": 0.00018896570909947475, |
| "loss": 0.4634, |
| "step": 120 |
| }, |
| { |
| "epoch": 0.20358306188925082, |
| "grad_norm": 1.2308176755905151, |
| "learning_rate": 0.00018770325650740345, |
| "loss": 0.6793, |
| "step": 125 |
| }, |
| { |
| "epoch": 0.21172638436482086, |
| "grad_norm": 1.0796643495559692, |
| "learning_rate": 0.00018637714018370253, |
| "loss": 0.4443, |
| "step": 130 |
| }, |
| { |
| "epoch": 0.21986970684039087, |
| "grad_norm": 0.6986652612686157, |
| "learning_rate": 0.00018498832275526988, |
| "loss": 0.3816, |
| "step": 135 |
| }, |
| { |
| "epoch": 0.2280130293159609, |
| "grad_norm": 1.1987988948822021, |
| "learning_rate": 0.00018353781236369064, |
| "loss": 0.5302, |
| "step": 140 |
| }, |
| { |
| "epoch": 0.23615635179153094, |
| "grad_norm": 1.2463840246200562, |
| "learning_rate": 0.00018202666193342833, |
| "loss": 0.6081, |
| "step": 145 |
| }, |
| { |
| "epoch": 0.24429967426710097, |
| "grad_norm": 1.1405540704727173, |
| "learning_rate": 0.00018045596840750723, |
| "loss": 0.5229, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.24429967426710097, |
| "eval_loss": 0.39218127727508545, |
| "eval_runtime": 82.8535, |
| "eval_samples_per_second": 3.042, |
| "eval_steps_per_second": 1.521, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.252442996742671, |
| "grad_norm": 1.1585423946380615, |
| "learning_rate": 0.0001788268719512427, |
| "loss": 0.425, |
| "step": 155 |
| }, |
| { |
| "epoch": 0.26058631921824105, |
| "grad_norm": 1.1226191520690918, |
| "learning_rate": 0.00017714055512459565, |
| "loss": 0.3537, |
| "step": 160 |
| }, |
| { |
| "epoch": 0.2687296416938111, |
| "grad_norm": 1.1310710906982422, |
| "learning_rate": 0.0001753982420237533, |
| "loss": 0.5322, |
| "step": 165 |
| }, |
| { |
| "epoch": 0.2768729641693811, |
| "grad_norm": 1.3631582260131836, |
| "learning_rate": 0.00017360119739255852, |
| "loss": 0.5435, |
| "step": 170 |
| }, |
| { |
| "epoch": 0.28501628664495116, |
| "grad_norm": 0.9981521964073181, |
| "learning_rate": 0.00017175072570443312, |
| "loss": 0.5163, |
| "step": 175 |
| }, |
| { |
| "epoch": 0.2931596091205212, |
| "grad_norm": 1.0105143785476685, |
| "learning_rate": 0.00016984817021546177, |
| "loss": 0.4881, |
| "step": 180 |
| }, |
| { |
| "epoch": 0.30130293159609123, |
| "grad_norm": 1.001007080078125, |
| "learning_rate": 0.000167894911989323, |
| "loss": 0.4659, |
| "step": 185 |
| }, |
| { |
| "epoch": 0.30944625407166126, |
| "grad_norm": 1.0138651132583618, |
| "learning_rate": 0.00016589236889477646, |
| "loss": 0.4601, |
| "step": 190 |
| }, |
| { |
| "epoch": 0.31758957654723124, |
| "grad_norm": 1.0092806816101074, |
| "learning_rate": 0.00016384199457643262, |
| "loss": 0.4872, |
| "step": 195 |
| }, |
| { |
| "epoch": 0.3257328990228013, |
| "grad_norm": 1.0924715995788574, |
| "learning_rate": 0.00016174527739955342, |
| "loss": 0.4727, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.3257328990228013, |
| "eval_loss": 0.3742806315422058, |
| "eval_runtime": 80.9683, |
| "eval_samples_per_second": 3.112, |
| "eval_steps_per_second": 1.556, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.3338762214983713, |
| "grad_norm": 1.0657118558883667, |
| "learning_rate": 0.00015960373936964892, |
| "loss": 0.4947, |
| "step": 205 |
| }, |
| { |
| "epoch": 0.34201954397394135, |
| "grad_norm": 1.0220006704330444, |
| "learning_rate": 0.0001574189350276545, |
| "loss": 0.4654, |
| "step": 210 |
| }, |
| { |
| "epoch": 0.3501628664495114, |
| "grad_norm": 1.0166069269180298, |
| "learning_rate": 0.00015519245032149083, |
| "loss": 0.4077, |
| "step": 215 |
| }, |
| { |
| "epoch": 0.3583061889250814, |
| "grad_norm": 0.9999421238899231, |
| "learning_rate": 0.0001529259014548253, |
| "loss": 0.5919, |
| "step": 220 |
| }, |
| { |
| "epoch": 0.36644951140065146, |
| "grad_norm": 1.1596579551696777, |
| "learning_rate": 0.00015062093371387097, |
| "loss": 0.5562, |
| "step": 225 |
| }, |
| { |
| "epoch": 0.3745928338762215, |
| "grad_norm": 0.9253465533256531, |
| "learning_rate": 0.00014827922027307451, |
| "loss": 0.4726, |
| "step": 230 |
| }, |
| { |
| "epoch": 0.38273615635179153, |
| "grad_norm": 1.1627142429351807, |
| "learning_rate": 0.00014590246098055996, |
| "loss": 0.5669, |
| "step": 235 |
| }, |
| { |
| "epoch": 0.39087947882736157, |
| "grad_norm": 1.0187082290649414, |
| "learning_rate": 0.00014349238112421024, |
| "loss": 0.3309, |
| "step": 240 |
| }, |
| { |
| "epoch": 0.3990228013029316, |
| "grad_norm": 1.183526635169983, |
| "learning_rate": 0.000141050730179282, |
| "loss": 0.5376, |
| "step": 245 |
| }, |
| { |
| "epoch": 0.40716612377850164, |
| "grad_norm": 1.2337042093276978, |
| "learning_rate": 0.0001385792805384625, |
| "loss": 0.5816, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.40716612377850164, |
| "eval_loss": 0.3654971122741699, |
| "eval_runtime": 85.0566, |
| "eval_samples_per_second": 2.963, |
| "eval_steps_per_second": 1.481, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.4153094462540717, |
| "grad_norm": 1.1129777431488037, |
| "learning_rate": 0.00013607982622529133, |
| "loss": 0.3451, |
| "step": 255 |
| }, |
| { |
| "epoch": 0.4234527687296417, |
| "grad_norm": 1.0514991283416748, |
| "learning_rate": 0.00013355418159187985, |
| "loss": 0.5078, |
| "step": 260 |
| }, |
| { |
| "epoch": 0.43159609120521175, |
| "grad_norm": 0.9415754675865173, |
| "learning_rate": 0.00013100418000187419, |
| "loss": 0.4047, |
| "step": 265 |
| }, |
| { |
| "epoch": 0.43973941368078173, |
| "grad_norm": 1.071070909500122, |
| "learning_rate": 0.0001284316724996181, |
| "loss": 0.6419, |
| "step": 270 |
| }, |
| { |
| "epoch": 0.44788273615635177, |
| "grad_norm": 1.2659767866134644, |
| "learning_rate": 0.00012583852646648095, |
| "loss": 0.4321, |
| "step": 275 |
| }, |
| { |
| "epoch": 0.4560260586319218, |
| "grad_norm": 0.9135441780090332, |
| "learning_rate": 0.00012322662426532708, |
| "loss": 0.4791, |
| "step": 280 |
| }, |
| { |
| "epoch": 0.46416938110749184, |
| "grad_norm": 0.9390333890914917, |
| "learning_rate": 0.00012059786187410984, |
| "loss": 0.4663, |
| "step": 285 |
| }, |
| { |
| "epoch": 0.4723127035830619, |
| "grad_norm": 1.0472079515457153, |
| "learning_rate": 0.00011795414750958265, |
| "loss": 0.4207, |
| "step": 290 |
| }, |
| { |
| "epoch": 0.4804560260586319, |
| "grad_norm": 0.923943042755127, |
| "learning_rate": 0.00011529740024212565, |
| "loss": 0.3595, |
| "step": 295 |
| }, |
| { |
| "epoch": 0.48859934853420195, |
| "grad_norm": 1.265942096710205, |
| "learning_rate": 0.00011262954860269399, |
| "loss": 0.5664, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48859934853420195, |
| "eval_loss": 0.35712096095085144, |
| "eval_runtime": 81.0935, |
| "eval_samples_per_second": 3.108, |
| "eval_steps_per_second": 1.554, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.496742671009772, |
| "grad_norm": 0.6245908141136169, |
| "learning_rate": 0.0001099525291828986, |
| "loss": 0.2506, |
| "step": 305 |
| }, |
| { |
| "epoch": 0.504885993485342, |
| "grad_norm": 1.0387402772903442, |
| "learning_rate": 0.00010726828522923562, |
| "loss": 0.3607, |
| "step": 310 |
| }, |
| { |
| "epoch": 0.5130293159609121, |
| "grad_norm": 1.0079439878463745, |
| "learning_rate": 0.00010457876523248518, |
| "loss": 0.4919, |
| "step": 315 |
| }, |
| { |
| "epoch": 0.5211726384364821, |
| "grad_norm": 1.384364128112793, |
| "learning_rate": 0.00010188592151330343, |
| "loss": 0.4161, |
| "step": 320 |
| }, |
| { |
| "epoch": 0.5293159609120521, |
| "grad_norm": 1.0133496522903442, |
| "learning_rate": 9.919170880503415e-05, |
| "loss": 0.5675, |
| "step": 325 |
| }, |
| { |
| "epoch": 0.5374592833876222, |
| "grad_norm": 1.0509177446365356, |
| "learning_rate": 9.649808283476941e-05, |
| "loss": 0.5886, |
| "step": 330 |
| }, |
| { |
| "epoch": 0.5456026058631922, |
| "grad_norm": 0.9355628490447998, |
| "learning_rate": 9.38069989036886e-05, |
| "loss": 0.4121, |
| "step": 335 |
| }, |
| { |
| "epoch": 0.5537459283387622, |
| "grad_norm": 0.9302476644515991, |
| "learning_rate": 9.112041046770653e-05, |
| "loss": 0.3472, |
| "step": 340 |
| }, |
| { |
| "epoch": 0.5618892508143323, |
| "grad_norm": 0.7449364066123962, |
| "learning_rate": 8.844026771946147e-05, |
| "loss": 0.3783, |
| "step": 345 |
| }, |
| { |
| "epoch": 0.5700325732899023, |
| "grad_norm": 0.9528508186340332, |
| "learning_rate": 8.57685161726715e-05, |
| "loss": 0.4806, |
| "step": 350 |
| }, |
| { |
| "epoch": 0.5700325732899023, |
| "eval_loss": 0.3507095277309418, |
| "eval_runtime": 79.3146, |
| "eval_samples_per_second": 3.177, |
| "eval_steps_per_second": 1.589, |
| "step": 350 |
| }, |
| { |
| "epoch": 0.5781758957654723, |
| "grad_norm": 0.8920520544052124, |
| "learning_rate": 8.310709524988743e-05, |
| "loss": 0.3767, |
| "step": 355 |
| }, |
| { |
| "epoch": 0.5863192182410424, |
| "grad_norm": 1.0878592729568481, |
| "learning_rate": 8.045793687466757e-05, |
| "loss": 0.3621, |
| "step": 360 |
| }, |
| { |
| "epoch": 0.5944625407166124, |
| "grad_norm": 0.582628607749939, |
| "learning_rate": 7.782296406919557e-05, |
| "loss": 0.442, |
| "step": 365 |
| }, |
| { |
| "epoch": 0.6026058631921825, |
| "grad_norm": 1.13759446144104, |
| "learning_rate": 7.520408955836007e-05, |
| "loss": 0.5035, |
| "step": 370 |
| }, |
| { |
| "epoch": 0.6107491856677525, |
| "grad_norm": 0.9254467487335205, |
| "learning_rate": 7.260321438130913e-05, |
| "loss": 0.3845, |
| "step": 375 |
| }, |
| { |
| "epoch": 0.6188925081433225, |
| "grad_norm": 0.834490954875946, |
| "learning_rate": 7.002222651148714e-05, |
| "loss": 0.3713, |
| "step": 380 |
| }, |
| { |
| "epoch": 0.6270358306188925, |
| "grad_norm": 1.0162910223007202, |
| "learning_rate": 6.746299948615631e-05, |
| "loss": 0.4964, |
| "step": 385 |
| }, |
| { |
| "epoch": 0.6351791530944625, |
| "grad_norm": 0.9034905433654785, |
| "learning_rate": 6.492739104639727e-05, |
| "loss": 0.4315, |
| "step": 390 |
| }, |
| { |
| "epoch": 0.6433224755700325, |
| "grad_norm": 1.0839813947677612, |
| "learning_rate": 6.24172417885762e-05, |
| "loss": 0.4281, |
| "step": 395 |
| }, |
| { |
| "epoch": 0.6514657980456026, |
| "grad_norm": 0.861652672290802, |
| "learning_rate": 5.9934373828257106e-05, |
| "loss": 0.3687, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.6514657980456026, |
| "eval_loss": 0.34634968638420105, |
| "eval_runtime": 78.9991, |
| "eval_samples_per_second": 3.19, |
| "eval_steps_per_second": 1.595, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.6596091205211726, |
| "grad_norm": 0.6291548609733582, |
| "learning_rate": 5.7480589477529545e-05, |
| "loss": 0.286, |
| "step": 405 |
| }, |
| { |
| "epoch": 0.6677524429967426, |
| "grad_norm": 0.9349259734153748, |
| "learning_rate": 5.505766993671162e-05, |
| "loss": 0.4552, |
| "step": 410 |
| }, |
| { |
| "epoch": 0.6758957654723127, |
| "grad_norm": 0.7627623081207275, |
| "learning_rate": 5.266737400137765e-05, |
| "loss": 0.4689, |
| "step": 415 |
| }, |
| { |
| "epoch": 0.6840390879478827, |
| "grad_norm": 0.8197122812271118, |
| "learning_rate": 5.031143678565005e-05, |
| "loss": 0.4964, |
| "step": 420 |
| }, |
| { |
| "epoch": 0.6921824104234527, |
| "grad_norm": 1.1312295198440552, |
| "learning_rate": 4.799156846268095e-05, |
| "loss": 0.4795, |
| "step": 425 |
| }, |
| { |
| "epoch": 0.7003257328990228, |
| "grad_norm": 0.7886964082717896, |
| "learning_rate": 4.57094530232389e-05, |
| "loss": 0.4403, |
| "step": 430 |
| }, |
| { |
| "epoch": 0.7084690553745928, |
| "grad_norm": 0.9440180063247681, |
| "learning_rate": 4.346674705330117e-05, |
| "loss": 0.3939, |
| "step": 435 |
| }, |
| { |
| "epoch": 0.7166123778501629, |
| "grad_norm": 1.0344818830490112, |
| "learning_rate": 4.1265078531538916e-05, |
| "loss": 0.3908, |
| "step": 440 |
| }, |
| { |
| "epoch": 0.7247557003257329, |
| "grad_norm": 0.7322747707366943, |
| "learning_rate": 3.9106045647569e-05, |
| "loss": 0.5379, |
| "step": 445 |
| }, |
| { |
| "epoch": 0.7328990228013029, |
| "grad_norm": 1.0375081300735474, |
| "learning_rate": 3.69912156418289e-05, |
| "loss": 0.5235, |
| "step": 450 |
| }, |
| { |
| "epoch": 0.7328990228013029, |
| "eval_loss": 0.3418995141983032, |
| "eval_runtime": 79.6011, |
| "eval_samples_per_second": 3.166, |
| "eval_steps_per_second": 1.583, |
| "step": 450 |
| }, |
| { |
| "epoch": 0.741042345276873, |
| "grad_norm": 1.0613162517547607, |
| "learning_rate": 3.492212366791831e-05, |
| "loss": 0.4515, |
| "step": 455 |
| }, |
| { |
| "epoch": 0.749185667752443, |
| "grad_norm": 0.9459342956542969, |
| "learning_rate": 3.290027167823204e-05, |
| "loss": 0.3406, |
| "step": 460 |
| }, |
| { |
| "epoch": 0.757328990228013, |
| "grad_norm": 1.2217379808425903, |
| "learning_rate": 3.092712733369387e-05, |
| "loss": 0.564, |
| "step": 465 |
| }, |
| { |
| "epoch": 0.7654723127035831, |
| "grad_norm": 1.1677855253219604, |
| "learning_rate": 2.9004122938382617e-05, |
| "loss": 0.6134, |
| "step": 470 |
| }, |
| { |
| "epoch": 0.7736156351791531, |
| "grad_norm": 0.9157093167304993, |
| "learning_rate": 2.7132654399823444e-05, |
| "loss": 0.5363, |
| "step": 475 |
| }, |
| { |
| "epoch": 0.7817589576547231, |
| "grad_norm": 1.1068150997161865, |
| "learning_rate": 2.531408021569982e-05, |
| "loss": 0.403, |
| "step": 480 |
| }, |
| { |
| "epoch": 0.7899022801302932, |
| "grad_norm": 0.9987042546272278, |
| "learning_rate": 2.3549720487720738e-05, |
| "loss": 0.4282, |
| "step": 485 |
| }, |
| { |
| "epoch": 0.7980456026058632, |
| "grad_norm": 0.940436601638794, |
| "learning_rate": 2.184085596336011e-05, |
| "loss": 0.4264, |
| "step": 490 |
| }, |
| { |
| "epoch": 0.8061889250814332, |
| "grad_norm": 0.5704993009567261, |
| "learning_rate": 2.0188727106162874e-05, |
| "loss": 0.2974, |
| "step": 495 |
| }, |
| { |
| "epoch": 0.8143322475570033, |
| "grad_norm": 0.8623107671737671, |
| "learning_rate": 1.8594533195293427e-05, |
| "loss": 0.3239, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8143322475570033, |
| "eval_loss": 0.33896583318710327, |
| "eval_runtime": 79.4278, |
| "eval_samples_per_second": 3.173, |
| "eval_steps_per_second": 1.586, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8224755700325733, |
| "grad_norm": 1.2364838123321533, |
| "learning_rate": 1.7059431454979824e-05, |
| "loss": 0.5516, |
| "step": 505 |
| }, |
| { |
| "epoch": 0.8306188925081434, |
| "grad_norm": 1.041014552116394, |
| "learning_rate": 1.5584536214485457e-05, |
| "loss": 0.4447, |
| "step": 510 |
| }, |
| { |
| "epoch": 0.8387622149837134, |
| "grad_norm": 1.2267398834228516, |
| "learning_rate": 1.4170918099218166e-05, |
| "loss": 0.5275, |
| "step": 515 |
| }, |
| { |
| "epoch": 0.8469055374592834, |
| "grad_norm": 0.5176966786384583, |
| "learning_rate": 1.2819603253564205e-05, |
| "loss": 0.4667, |
| "step": 520 |
| }, |
| { |
| "epoch": 0.8550488599348535, |
| "grad_norm": 1.1171722412109375, |
| "learning_rate": 1.153157259601062e-05, |
| "loss": 0.5848, |
| "step": 525 |
| }, |
| { |
| "epoch": 0.8631921824104235, |
| "grad_norm": 1.046492099761963, |
| "learning_rate": 1.030776110709718e-05, |
| "loss": 0.3975, |
| "step": 530 |
| }, |
| { |
| "epoch": 0.8713355048859935, |
| "grad_norm": 1.046989917755127, |
| "learning_rate": 9.149057150714801e-06, |
| "loss": 0.4733, |
| "step": 535 |
| }, |
| { |
| "epoch": 0.8794788273615635, |
| "grad_norm": 0.9033722281455994, |
| "learning_rate": 8.056301829242784e-06, |
| "loss": 0.4767, |
| "step": 540 |
| }, |
| { |
| "epoch": 0.8876221498371335, |
| "grad_norm": 0.8106915354728699, |
| "learning_rate": 7.030288372993066e-06, |
| "loss": 0.6518, |
| "step": 545 |
| }, |
| { |
| "epoch": 0.8957654723127035, |
| "grad_norm": 0.7800573110580444, |
| "learning_rate": 6.0717615644051206e-06, |
| "loss": 0.3643, |
| "step": 550 |
| }, |
| { |
| "epoch": 0.8957654723127035, |
| "eval_loss": 0.3375564515590668, |
| "eval_runtime": 78.9411, |
| "eval_samples_per_second": 3.192, |
| "eval_steps_per_second": 1.596, |
| "step": 550 |
| }, |
| { |
| "epoch": 0.9039087947882736, |
| "grad_norm": 0.7835620641708374, |
| "learning_rate": 5.181417197408734e-06, |
| "loss": 0.3776, |
| "step": 555 |
| }, |
| { |
| "epoch": 0.9120521172638436, |
| "grad_norm": 0.9366523623466492, |
| "learning_rate": 4.359901572347758e-06, |
| "loss": 0.5273, |
| "step": 560 |
| }, |
| { |
| "epoch": 0.9201954397394136, |
| "grad_norm": 0.9118573069572449, |
| "learning_rate": 3.607811026831176e-06, |
| "loss": 0.4158, |
| "step": 565 |
| }, |
| { |
| "epoch": 0.9283387622149837, |
| "grad_norm": 1.0393532514572144, |
| "learning_rate": 2.9256915028519573e-06, |
| "loss": 0.4686, |
| "step": 570 |
| }, |
| { |
| "epoch": 0.9364820846905537, |
| "grad_norm": 0.9593573212623596, |
| "learning_rate": 2.3140381504882737e-06, |
| "loss": 0.5106, |
| "step": 575 |
| }, |
| { |
| "epoch": 0.9446254071661238, |
| "grad_norm": 1.3818498849868774, |
| "learning_rate": 1.7732949684743594e-06, |
| "loss": 0.5435, |
| "step": 580 |
| }, |
| { |
| "epoch": 0.9527687296416938, |
| "grad_norm": 1.069556713104248, |
| "learning_rate": 1.303854481902067e-06, |
| "loss": 0.3819, |
| "step": 585 |
| }, |
| { |
| "epoch": 0.9609120521172638, |
| "grad_norm": 0.6016046404838562, |
| "learning_rate": 9.060574572873237e-07, |
| "loss": 0.3293, |
| "step": 590 |
| }, |
| { |
| "epoch": 0.9690553745928339, |
| "grad_norm": 0.6195086240768433, |
| "learning_rate": 5.801926552078563e-07, |
| "loss": 0.3587, |
| "step": 595 |
| }, |
| { |
| "epoch": 0.9771986970684039, |
| "grad_norm": 0.9202044010162354, |
| "learning_rate": 3.264966206921294e-07, |
| "loss": 0.3949, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.9771986970684039, |
| "eval_loss": 0.33704105019569397, |
| "eval_runtime": 79.1876, |
| "eval_samples_per_second": 3.182, |
| "eval_steps_per_second": 1.591, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.9853420195439739, |
| "grad_norm": 0.944168210029602, |
| "learning_rate": 1.451535115114866e-07, |
| "loss": 0.2936, |
| "step": 605 |
| }, |
| { |
| "epoch": 0.993485342019544, |
| "grad_norm": 1.092924952507019, |
| "learning_rate": 3.629496450011649e-08, |
| "loss": 0.3938, |
| "step": 610 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 614, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 1, |
| "save_steps": 100, |
| "stateful_callbacks": { |
| "EarlyStoppingCallback": { |
| "args": { |
| "early_stopping_patience": 2, |
| "early_stopping_threshold": 0.001 |
| }, |
| "attributes": { |
| "early_stopping_patience_counter": 1 |
| } |
| }, |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 2.5087716731374203e+18, |
| "train_batch_size": 2, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
