| { |
| "best_global_step": 600, |
| "best_metric": 0.266491562128067, |
| "best_model_checkpoint": "/workspace/anchor/qwen-compassion-web-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": 8.798036575317383, |
| "learning_rate": 2.5806451612903226e-05, |
| "loss": 1.2046, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.016286644951140065, |
| "grad_norm": 3.6134917736053467, |
| "learning_rate": 5.8064516129032266e-05, |
| "loss": 0.941, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.024429967426710098, |
| "grad_norm": 2.604928731918335, |
| "learning_rate": 9.032258064516129e-05, |
| "loss": 0.8643, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.03257328990228013, |
| "grad_norm": 1.981958270072937, |
| "learning_rate": 0.00012258064516129034, |
| "loss": 0.8232, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.04071661237785016, |
| "grad_norm": 3.514120101928711, |
| "learning_rate": 0.00015483870967741937, |
| "loss": 0.7627, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.048859934853420196, |
| "grad_norm": 2.8703806400299072, |
| "learning_rate": 0.0001870967741935484, |
| "loss": 0.604, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.057003257328990226, |
| "grad_norm": 2.296351194381714, |
| "learning_rate": 0.00019998693330688282, |
| "loss": 0.6969, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.06514657980456026, |
| "grad_norm": 2.8257081508636475, |
| "learning_rate": 0.00019990709365929677, |
| "loss": 0.5599, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.0732899022801303, |
| "grad_norm": 2.1461329460144043, |
| "learning_rate": 0.00019975473152335726, |
| "loss": 0.5626, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.08143322475570032, |
| "grad_norm": 1.7384521961212158, |
| "learning_rate": 0.0001995299574986306, |
| "loss": 0.5407, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.08143322475570032, |
| "eval_loss": 0.3758981227874756, |
| "eval_runtime": 194.5072, |
| "eval_samples_per_second": 1.296, |
| "eval_steps_per_second": 0.648, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.08957654723127036, |
| "grad_norm": 1.4657418727874756, |
| "learning_rate": 0.00019923293474842174, |
| "loss": 0.5894, |
| "step": 55 |
| }, |
| { |
| "epoch": 0.09771986970684039, |
| "grad_norm": 1.2558283805847168, |
| "learning_rate": 0.00019886387888133413, |
| "loss": 0.4949, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.10586319218241043, |
| "grad_norm": 1.2131683826446533, |
| "learning_rate": 0.00019842305779475968, |
| "loss": 0.3924, |
| "step": 65 |
| }, |
| { |
| "epoch": 0.11400651465798045, |
| "grad_norm": 1.3360671997070312, |
| "learning_rate": 0.0001979107914804122, |
| "loss": 0.5551, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.12214983713355049, |
| "grad_norm": 0.9418010711669922, |
| "learning_rate": 0.00019732745179204552, |
| "loss": 0.4046, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.13029315960912052, |
| "grad_norm": 1.7212756872177124, |
| "learning_rate": 0.00019667346217552527, |
| "loss": 0.4057, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.13843648208469056, |
| "grad_norm": 1.399787425994873, |
| "learning_rate": 0.00019594929736144976, |
| "loss": 0.5889, |
| "step": 85 |
| }, |
| { |
| "epoch": 0.1465798045602606, |
| "grad_norm": 1.2302014827728271, |
| "learning_rate": 0.00019515548302054335, |
| "loss": 0.4786, |
| "step": 90 |
| }, |
| { |
| "epoch": 0.15472312703583063, |
| "grad_norm": 1.103279948234558, |
| "learning_rate": 0.0001942925953820725, |
| "loss": 0.4461, |
| "step": 95 |
| }, |
| { |
| "epoch": 0.16286644951140064, |
| "grad_norm": 1.1707514524459839, |
| "learning_rate": 0.00019336126081556134, |
| "loss": 0.5073, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16286644951140064, |
| "eval_loss": 0.3113905191421509, |
| "eval_runtime": 84.1681, |
| "eval_samples_per_second": 2.994, |
| "eval_steps_per_second": 1.497, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.17100977198697068, |
| "grad_norm": 1.198279619216919, |
| "learning_rate": 0.00019236215537611046, |
| "loss": 0.4347, |
| "step": 105 |
| }, |
| { |
| "epoch": 0.1791530944625407, |
| "grad_norm": 1.2491925954818726, |
| "learning_rate": 0.00019129600431364897, |
| "loss": 0.455, |
| "step": 110 |
| }, |
| { |
| "epoch": 0.18729641693811075, |
| "grad_norm": 1.2491847276687622, |
| "learning_rate": 0.00019016358154647618, |
| "loss": 0.4474, |
| "step": 115 |
| }, |
| { |
| "epoch": 0.19543973941368079, |
| "grad_norm": 1.3569931983947754, |
| "learning_rate": 0.00018896570909947475, |
| "loss": 0.4234, |
| "step": 120 |
| }, |
| { |
| "epoch": 0.20358306188925082, |
| "grad_norm": 1.1421902179718018, |
| "learning_rate": 0.00018770325650740345, |
| "loss": 0.4171, |
| "step": 125 |
| }, |
| { |
| "epoch": 0.21172638436482086, |
| "grad_norm": 1.1854735612869263, |
| "learning_rate": 0.00018637714018370253, |
| "loss": 0.476, |
| "step": 130 |
| }, |
| { |
| "epoch": 0.21986970684039087, |
| "grad_norm": 0.8009733557701111, |
| "learning_rate": 0.00018498832275526988, |
| "loss": 0.4296, |
| "step": 135 |
| }, |
| { |
| "epoch": 0.2280130293159609, |
| "grad_norm": 0.690112829208374, |
| "learning_rate": 0.00018353781236369064, |
| "loss": 0.4327, |
| "step": 140 |
| }, |
| { |
| "epoch": 0.23615635179153094, |
| "grad_norm": 1.194413185119629, |
| "learning_rate": 0.00018202666193342833, |
| "loss": 0.4803, |
| "step": 145 |
| }, |
| { |
| "epoch": 0.24429967426710097, |
| "grad_norm": 1.320000171661377, |
| "learning_rate": 0.00018045596840750723, |
| "loss": 0.4782, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.24429967426710097, |
| "eval_loss": 0.29872405529022217, |
| "eval_runtime": 83.9329, |
| "eval_samples_per_second": 3.002, |
| "eval_steps_per_second": 1.501, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.252442996742671, |
| "grad_norm": 0.946649968624115, |
| "learning_rate": 0.0001788268719512427, |
| "loss": 0.4411, |
| "step": 155 |
| }, |
| { |
| "epoch": 0.26058631921824105, |
| "grad_norm": 1.3469047546386719, |
| "learning_rate": 0.00017714055512459565, |
| "loss": 0.4299, |
| "step": 160 |
| }, |
| { |
| "epoch": 0.2687296416938111, |
| "grad_norm": 0.9297264218330383, |
| "learning_rate": 0.0001753982420237533, |
| "loss": 0.5015, |
| "step": 165 |
| }, |
| { |
| "epoch": 0.2768729641693811, |
| "grad_norm": 1.2105650901794434, |
| "learning_rate": 0.00017360119739255852, |
| "loss": 0.3346, |
| "step": 170 |
| }, |
| { |
| "epoch": 0.28501628664495116, |
| "grad_norm": 0.8039602041244507, |
| "learning_rate": 0.00017175072570443312, |
| "loss": 0.4798, |
| "step": 175 |
| }, |
| { |
| "epoch": 0.2931596091205212, |
| "grad_norm": 0.7823976874351501, |
| "learning_rate": 0.00016984817021546177, |
| "loss": 0.3705, |
| "step": 180 |
| }, |
| { |
| "epoch": 0.30130293159609123, |
| "grad_norm": 1.5801986455917358, |
| "learning_rate": 0.000167894911989323, |
| "loss": 0.4246, |
| "step": 185 |
| }, |
| { |
| "epoch": 0.30944625407166126, |
| "grad_norm": 1.122887372970581, |
| "learning_rate": 0.00016589236889477646, |
| "loss": 0.3341, |
| "step": 190 |
| }, |
| { |
| "epoch": 0.31758957654723124, |
| "grad_norm": 1.2335751056671143, |
| "learning_rate": 0.00016384199457643262, |
| "loss": 0.4647, |
| "step": 195 |
| }, |
| { |
| "epoch": 0.3257328990228013, |
| "grad_norm": 1.0149917602539062, |
| "learning_rate": 0.00016174527739955342, |
| "loss": 0.5815, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.3257328990228013, |
| "eval_loss": 0.2909018397331238, |
| "eval_runtime": 84.2104, |
| "eval_samples_per_second": 2.993, |
| "eval_steps_per_second": 1.496, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.3338762214983713, |
| "grad_norm": 0.6360901594161987, |
| "learning_rate": 0.00015960373936964892, |
| "loss": 0.3832, |
| "step": 205 |
| }, |
| { |
| "epoch": 0.34201954397394135, |
| "grad_norm": 1.1839494705200195, |
| "learning_rate": 0.0001574189350276545, |
| "loss": 0.4736, |
| "step": 210 |
| }, |
| { |
| "epoch": 0.3501628664495114, |
| "grad_norm": 1.2935487031936646, |
| "learning_rate": 0.00015519245032149083, |
| "loss": 0.5206, |
| "step": 215 |
| }, |
| { |
| "epoch": 0.3583061889250814, |
| "grad_norm": 1.0064383745193481, |
| "learning_rate": 0.0001529259014548253, |
| "loss": 0.4329, |
| "step": 220 |
| }, |
| { |
| "epoch": 0.36644951140065146, |
| "grad_norm": 1.1821966171264648, |
| "learning_rate": 0.00015062093371387097, |
| "loss": 0.4376, |
| "step": 225 |
| }, |
| { |
| "epoch": 0.3745928338762215, |
| "grad_norm": 1.3237477540969849, |
| "learning_rate": 0.00014827922027307451, |
| "loss": 0.4052, |
| "step": 230 |
| }, |
| { |
| "epoch": 0.38273615635179153, |
| "grad_norm": 1.0191577672958374, |
| "learning_rate": 0.00014590246098055996, |
| "loss": 0.4628, |
| "step": 235 |
| }, |
| { |
| "epoch": 0.39087947882736157, |
| "grad_norm": 0.8374785780906677, |
| "learning_rate": 0.00014349238112421024, |
| "loss": 0.3099, |
| "step": 240 |
| }, |
| { |
| "epoch": 0.3990228013029316, |
| "grad_norm": 0.9581987857818604, |
| "learning_rate": 0.000141050730179282, |
| "loss": 0.4072, |
| "step": 245 |
| }, |
| { |
| "epoch": 0.40716612377850164, |
| "grad_norm": 1.1733781099319458, |
| "learning_rate": 0.0001385792805384625, |
| "loss": 0.5048, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.40716612377850164, |
| "eval_loss": 0.2858830988407135, |
| "eval_runtime": 84.1543, |
| "eval_samples_per_second": 2.995, |
| "eval_steps_per_second": 1.497, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.4153094462540717, |
| "grad_norm": 1.1163829565048218, |
| "learning_rate": 0.00013607982622529133, |
| "loss": 0.41, |
| "step": 255 |
| }, |
| { |
| "epoch": 0.4234527687296417, |
| "grad_norm": 1.2686806917190552, |
| "learning_rate": 0.00013355418159187985, |
| "loss": 0.4759, |
| "step": 260 |
| }, |
| { |
| "epoch": 0.43159609120521175, |
| "grad_norm": 0.7049103379249573, |
| "learning_rate": 0.00013100418000187419, |
| "loss": 0.46, |
| "step": 265 |
| }, |
| { |
| "epoch": 0.43973941368078173, |
| "grad_norm": 1.0488688945770264, |
| "learning_rate": 0.0001284316724996181, |
| "loss": 0.4648, |
| "step": 270 |
| }, |
| { |
| "epoch": 0.44788273615635177, |
| "grad_norm": 1.1495773792266846, |
| "learning_rate": 0.00012583852646648095, |
| "loss": 0.5741, |
| "step": 275 |
| }, |
| { |
| "epoch": 0.4560260586319218, |
| "grad_norm": 1.1209203004837036, |
| "learning_rate": 0.00012322662426532708, |
| "loss": 0.4615, |
| "step": 280 |
| }, |
| { |
| "epoch": 0.46416938110749184, |
| "grad_norm": 0.6924991011619568, |
| "learning_rate": 0.00012059786187410984, |
| "loss": 0.3035, |
| "step": 285 |
| }, |
| { |
| "epoch": 0.4723127035830619, |
| "grad_norm": 0.6520649790763855, |
| "learning_rate": 0.00011795414750958265, |
| "loss": 0.2829, |
| "step": 290 |
| }, |
| { |
| "epoch": 0.4804560260586319, |
| "grad_norm": 1.101690649986267, |
| "learning_rate": 0.00011529740024212565, |
| "loss": 0.4049, |
| "step": 295 |
| }, |
| { |
| "epoch": 0.48859934853420195, |
| "grad_norm": 1.2471281290054321, |
| "learning_rate": 0.00011262954860269399, |
| "loss": 0.5541, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48859934853420195, |
| "eval_loss": 0.2825718820095062, |
| "eval_runtime": 84.4271, |
| "eval_samples_per_second": 2.985, |
| "eval_steps_per_second": 1.492, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.496742671009772, |
| "grad_norm": 1.1756144762039185, |
| "learning_rate": 0.0001099525291828986, |
| "loss": 0.3676, |
| "step": 305 |
| }, |
| { |
| "epoch": 0.504885993485342, |
| "grad_norm": 1.3010112047195435, |
| "learning_rate": 0.00010726828522923562, |
| "loss": 0.4012, |
| "step": 310 |
| }, |
| { |
| "epoch": 0.5130293159609121, |
| "grad_norm": 1.1838953495025635, |
| "learning_rate": 0.00010457876523248518, |
| "loss": 0.4652, |
| "step": 315 |
| }, |
| { |
| "epoch": 0.5211726384364821, |
| "grad_norm": 0.6249113082885742, |
| "learning_rate": 0.00010188592151330343, |
| "loss": 0.3431, |
| "step": 320 |
| }, |
| { |
| "epoch": 0.5293159609120521, |
| "grad_norm": 0.6638064384460449, |
| "learning_rate": 9.919170880503415e-05, |
| "loss": 0.4083, |
| "step": 325 |
| }, |
| { |
| "epoch": 0.5374592833876222, |
| "grad_norm": 1.111833095550537, |
| "learning_rate": 9.649808283476941e-05, |
| "loss": 0.2453, |
| "step": 330 |
| }, |
| { |
| "epoch": 0.5456026058631922, |
| "grad_norm": 0.7372708320617676, |
| "learning_rate": 9.38069989036886e-05, |
| "loss": 0.4156, |
| "step": 335 |
| }, |
| { |
| "epoch": 0.5537459283387622, |
| "grad_norm": 0.7365127205848694, |
| "learning_rate": 9.112041046770653e-05, |
| "loss": 0.2237, |
| "step": 340 |
| }, |
| { |
| "epoch": 0.5618892508143323, |
| "grad_norm": 1.104100227355957, |
| "learning_rate": 8.844026771946147e-05, |
| "loss": 0.4237, |
| "step": 345 |
| }, |
| { |
| "epoch": 0.5700325732899023, |
| "grad_norm": 1.1901392936706543, |
| "learning_rate": 8.57685161726715e-05, |
| "loss": 0.4926, |
| "step": 350 |
| }, |
| { |
| "epoch": 0.5700325732899023, |
| "eval_loss": 0.276626855134964, |
| "eval_runtime": 83.3294, |
| "eval_samples_per_second": 3.024, |
| "eval_steps_per_second": 1.512, |
| "step": 350 |
| }, |
| { |
| "epoch": 0.5781758957654723, |
| "grad_norm": 0.9266198873519897, |
| "learning_rate": 8.310709524988743e-05, |
| "loss": 0.4329, |
| "step": 355 |
| }, |
| { |
| "epoch": 0.5863192182410424, |
| "grad_norm": 1.0464011430740356, |
| "learning_rate": 8.045793687466757e-05, |
| "loss": 0.3268, |
| "step": 360 |
| }, |
| { |
| "epoch": 0.5944625407166124, |
| "grad_norm": 1.0177321434020996, |
| "learning_rate": 7.782296406919557e-05, |
| "loss": 0.4688, |
| "step": 365 |
| }, |
| { |
| "epoch": 0.6026058631921825, |
| "grad_norm": 1.087843418121338, |
| "learning_rate": 7.520408955836007e-05, |
| "loss": 0.2971, |
| "step": 370 |
| }, |
| { |
| "epoch": 0.6107491856677525, |
| "grad_norm": 0.6417682766914368, |
| "learning_rate": 7.260321438130913e-05, |
| "loss": 0.3314, |
| "step": 375 |
| }, |
| { |
| "epoch": 0.6188925081433225, |
| "grad_norm": 0.8692435622215271, |
| "learning_rate": 7.002222651148714e-05, |
| "loss": 0.2523, |
| "step": 380 |
| }, |
| { |
| "epoch": 0.6270358306188925, |
| "grad_norm": 1.3478347063064575, |
| "learning_rate": 6.746299948615631e-05, |
| "loss": 0.5737, |
| "step": 385 |
| }, |
| { |
| "epoch": 0.6351791530944625, |
| "grad_norm": 0.9037352800369263, |
| "learning_rate": 6.492739104639727e-05, |
| "loss": 0.3183, |
| "step": 390 |
| }, |
| { |
| "epoch": 0.6433224755700325, |
| "grad_norm": 0.8303831219673157, |
| "learning_rate": 6.24172417885762e-05, |
| "loss": 0.4005, |
| "step": 395 |
| }, |
| { |
| "epoch": 0.6514657980456026, |
| "grad_norm": 1.0739305019378662, |
| "learning_rate": 5.9934373828257106e-05, |
| "loss": 0.3355, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.6514657980456026, |
| "eval_loss": 0.27336135506629944, |
| "eval_runtime": 84.9655, |
| "eval_samples_per_second": 2.966, |
| "eval_steps_per_second": 1.483, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.6596091205211726, |
| "grad_norm": 0.6205328702926636, |
| "learning_rate": 5.7480589477529545e-05, |
| "loss": 0.3713, |
| "step": 405 |
| }, |
| { |
| "epoch": 0.6677524429967426, |
| "grad_norm": 0.9618639349937439, |
| "learning_rate": 5.505766993671162e-05, |
| "loss": 0.4575, |
| "step": 410 |
| }, |
| { |
| "epoch": 0.6758957654723127, |
| "grad_norm": 1.1340993642807007, |
| "learning_rate": 5.266737400137765e-05, |
| "loss": 0.2969, |
| "step": 415 |
| }, |
| { |
| "epoch": 0.6840390879478827, |
| "grad_norm": 1.3060582876205444, |
| "learning_rate": 5.031143678565005e-05, |
| "loss": 0.338, |
| "step": 420 |
| }, |
| { |
| "epoch": 0.6921824104234527, |
| "grad_norm": 0.5971317887306213, |
| "learning_rate": 4.799156846268095e-05, |
| "loss": 0.3014, |
| "step": 425 |
| }, |
| { |
| "epoch": 0.7003257328990228, |
| "grad_norm": 1.2781586647033691, |
| "learning_rate": 4.57094530232389e-05, |
| "loss": 0.4378, |
| "step": 430 |
| }, |
| { |
| "epoch": 0.7084690553745928, |
| "grad_norm": 0.9771912693977356, |
| "learning_rate": 4.346674705330117e-05, |
| "loss": 0.4439, |
| "step": 435 |
| }, |
| { |
| "epoch": 0.7166123778501629, |
| "grad_norm": 1.0740609169006348, |
| "learning_rate": 4.1265078531538916e-05, |
| "loss": 0.3471, |
| "step": 440 |
| }, |
| { |
| "epoch": 0.7247557003257329, |
| "grad_norm": 1.1676493883132935, |
| "learning_rate": 3.9106045647569e-05, |
| "loss": 0.4143, |
| "step": 445 |
| }, |
| { |
| "epoch": 0.7328990228013029, |
| "grad_norm": 1.323346495628357, |
| "learning_rate": 3.69912156418289e-05, |
| "loss": 0.5194, |
| "step": 450 |
| }, |
| { |
| "epoch": 0.7328990228013029, |
| "eval_loss": 0.2706548273563385, |
| "eval_runtime": 82.9884, |
| "eval_samples_per_second": 3.037, |
| "eval_steps_per_second": 1.518, |
| "step": 450 |
| }, |
| { |
| "epoch": 0.741042345276873, |
| "grad_norm": 1.1108720302581787, |
| "learning_rate": 3.492212366791831e-05, |
| "loss": 0.3813, |
| "step": 455 |
| }, |
| { |
| "epoch": 0.749185667752443, |
| "grad_norm": 1.0551023483276367, |
| "learning_rate": 3.290027167823204e-05, |
| "loss": 0.3814, |
| "step": 460 |
| }, |
| { |
| "epoch": 0.757328990228013, |
| "grad_norm": 1.0584551095962524, |
| "learning_rate": 3.092712733369387e-05, |
| "loss": 0.3479, |
| "step": 465 |
| }, |
| { |
| "epoch": 0.7654723127035831, |
| "grad_norm": 0.723073422908783, |
| "learning_rate": 2.9004122938382617e-05, |
| "loss": 0.322, |
| "step": 470 |
| }, |
| { |
| "epoch": 0.7736156351791531, |
| "grad_norm": 0.9690468907356262, |
| "learning_rate": 2.7132654399823444e-05, |
| "loss": 0.4283, |
| "step": 475 |
| }, |
| { |
| "epoch": 0.7817589576547231, |
| "grad_norm": 0.8642581105232239, |
| "learning_rate": 2.531408021569982e-05, |
| "loss": 0.3914, |
| "step": 480 |
| }, |
| { |
| "epoch": 0.7899022801302932, |
| "grad_norm": 0.9394587278366089, |
| "learning_rate": 2.3549720487720738e-05, |
| "loss": 0.4853, |
| "step": 485 |
| }, |
| { |
| "epoch": 0.7980456026058632, |
| "grad_norm": 1.0716526508331299, |
| "learning_rate": 2.184085596336011e-05, |
| "loss": 0.4108, |
| "step": 490 |
| }, |
| { |
| "epoch": 0.8061889250814332, |
| "grad_norm": 1.1567680835723877, |
| "learning_rate": 2.0188727106162874e-05, |
| "loss": 0.4587, |
| "step": 495 |
| }, |
| { |
| "epoch": 0.8143322475570033, |
| "grad_norm": 0.9111871719360352, |
| "learning_rate": 1.8594533195293427e-05, |
| "loss": 0.4122, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8143322475570033, |
| "eval_loss": 0.2679363489151001, |
| "eval_runtime": 83.1827, |
| "eval_samples_per_second": 3.029, |
| "eval_steps_per_second": 1.515, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8224755700325733, |
| "grad_norm": 1.0076968669891357, |
| "learning_rate": 1.7059431454979824e-05, |
| "loss": 0.4203, |
| "step": 505 |
| }, |
| { |
| "epoch": 0.8306188925081434, |
| "grad_norm": 1.4316879510879517, |
| "learning_rate": 1.5584536214485457e-05, |
| "loss": 0.3873, |
| "step": 510 |
| }, |
| { |
| "epoch": 0.8387622149837134, |
| "grad_norm": 1.0346064567565918, |
| "learning_rate": 1.4170918099218166e-05, |
| "loss": 0.3811, |
| "step": 515 |
| }, |
| { |
| "epoch": 0.8469055374592834, |
| "grad_norm": 0.9672597646713257, |
| "learning_rate": 1.2819603253564205e-05, |
| "loss": 0.3536, |
| "step": 520 |
| }, |
| { |
| "epoch": 0.8550488599348535, |
| "grad_norm": 0.9941731691360474, |
| "learning_rate": 1.153157259601062e-05, |
| "loss": 0.3703, |
| "step": 525 |
| }, |
| { |
| "epoch": 0.8631921824104235, |
| "grad_norm": 0.9923365712165833, |
| "learning_rate": 1.030776110709718e-05, |
| "loss": 0.4362, |
| "step": 530 |
| }, |
| { |
| "epoch": 0.8713355048859935, |
| "grad_norm": 1.0414419174194336, |
| "learning_rate": 9.149057150714801e-06, |
| "loss": 0.4203, |
| "step": 535 |
| }, |
| { |
| "epoch": 0.8794788273615635, |
| "grad_norm": 1.0432775020599365, |
| "learning_rate": 8.056301829242784e-06, |
| "loss": 0.3753, |
| "step": 540 |
| }, |
| { |
| "epoch": 0.8876221498371335, |
| "grad_norm": 0.6528933048248291, |
| "learning_rate": 7.030288372993066e-06, |
| "loss": 0.2481, |
| "step": 545 |
| }, |
| { |
| "epoch": 0.8957654723127035, |
| "grad_norm": 1.0773026943206787, |
| "learning_rate": 6.0717615644051206e-06, |
| "loss": 0.4472, |
| "step": 550 |
| }, |
| { |
| "epoch": 0.8957654723127035, |
| "eval_loss": 0.26670876145362854, |
| "eval_runtime": 83.13, |
| "eval_samples_per_second": 3.031, |
| "eval_steps_per_second": 1.516, |
| "step": 550 |
| }, |
| { |
| "epoch": 0.9039087947882736, |
| "grad_norm": 0.8582516312599182, |
| "learning_rate": 5.181417197408734e-06, |
| "loss": 0.4482, |
| "step": 555 |
| }, |
| { |
| "epoch": 0.9120521172638436, |
| "grad_norm": 0.8720370531082153, |
| "learning_rate": 4.359901572347758e-06, |
| "loss": 0.3908, |
| "step": 560 |
| }, |
| { |
| "epoch": 0.9201954397394136, |
| "grad_norm": 1.0178829431533813, |
| "learning_rate": 3.607811026831176e-06, |
| "loss": 0.4675, |
| "step": 565 |
| }, |
| { |
| "epoch": 0.9283387622149837, |
| "grad_norm": 1.0477415323257446, |
| "learning_rate": 2.9256915028519573e-06, |
| "loss": 0.3136, |
| "step": 570 |
| }, |
| { |
| "epoch": 0.9364820846905537, |
| "grad_norm": 0.9757538437843323, |
| "learning_rate": 2.3140381504882737e-06, |
| "loss": 0.4209, |
| "step": 575 |
| }, |
| { |
| "epoch": 0.9446254071661238, |
| "grad_norm": 0.9059130549430847, |
| "learning_rate": 1.7732949684743594e-06, |
| "loss": 0.3305, |
| "step": 580 |
| }, |
| { |
| "epoch": 0.9527687296416938, |
| "grad_norm": 1.3074954748153687, |
| "learning_rate": 1.303854481902067e-06, |
| "loss": 0.3919, |
| "step": 585 |
| }, |
| { |
| "epoch": 0.9609120521172638, |
| "grad_norm": 1.0708932876586914, |
| "learning_rate": 9.060574572873237e-07, |
| "loss": 0.3902, |
| "step": 590 |
| }, |
| { |
| "epoch": 0.9690553745928339, |
| "grad_norm": 1.1010503768920898, |
| "learning_rate": 5.801926552078563e-07, |
| "loss": 0.3606, |
| "step": 595 |
| }, |
| { |
| "epoch": 0.9771986970684039, |
| "grad_norm": 1.148178219795227, |
| "learning_rate": 3.264966206921294e-07, |
| "loss": 0.389, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.9771986970684039, |
| "eval_loss": 0.266491562128067, |
| "eval_runtime": 83.2326, |
| "eval_samples_per_second": 3.028, |
| "eval_steps_per_second": 1.514, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.9853420195439739, |
| "grad_norm": 1.2450178861618042, |
| "learning_rate": 1.451535115114866e-07, |
| "loss": 0.4498, |
| "step": 605 |
| }, |
| { |
| "epoch": 0.993485342019544, |
| "grad_norm": 0.9275217652320862, |
| "learning_rate": 3.629496450011649e-08, |
| "loss": 0.3203, |
| "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": 2.5080656292484915e+18, |
| "train_batch_size": 2, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
