| { | |
| "best_metric": 0.19583497941493988, | |
| "best_model_checkpoint": "multilingual-e5-small-aligned-quality-20241214-new/checkpoint-23439", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 23439, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06399590426212723, | |
| "grad_norm": 1.3940926790237427, | |
| "learning_rate": 4.8933401595631215e-05, | |
| "loss": 0.3708, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.12799180852425446, | |
| "grad_norm": 1.3164088726043701, | |
| "learning_rate": 4.786680319126243e-05, | |
| "loss": 0.3132, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.19198771278638166, | |
| "grad_norm": 1.3515187501907349, | |
| "learning_rate": 4.680020478689364e-05, | |
| "loss": 0.2973, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.2559836170485089, | |
| "grad_norm": 1.7389737367630005, | |
| "learning_rate": 4.573360638252485e-05, | |
| "loss": 0.2877, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.3199795213106361, | |
| "grad_norm": 1.2677313089370728, | |
| "learning_rate": 4.4667007978156063e-05, | |
| "loss": 0.2803, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.3839754255727633, | |
| "grad_norm": 1.3927785158157349, | |
| "learning_rate": 4.360040957378728e-05, | |
| "loss": 0.2743, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.4479713298348906, | |
| "grad_norm": 1.628334879875183, | |
| "learning_rate": 4.2533811169418495e-05, | |
| "loss": 0.2747, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.5119672340970178, | |
| "grad_norm": 1.4009227752685547, | |
| "learning_rate": 4.146721276504971e-05, | |
| "loss": 0.2672, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.575963138359145, | |
| "grad_norm": 1.4522840976715088, | |
| "learning_rate": 4.040061436068092e-05, | |
| "loss": 0.264, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.6399590426212722, | |
| "grad_norm": 1.5413362979888916, | |
| "learning_rate": 3.933401595631213e-05, | |
| "loss": 0.2588, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.7039549468833994, | |
| "grad_norm": 1.2421332597732544, | |
| "learning_rate": 3.8267417551943344e-05, | |
| "loss": 0.253, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.7679508511455266, | |
| "grad_norm": 1.3509016036987305, | |
| "learning_rate": 3.7200819147574556e-05, | |
| "loss": 0.2539, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.831946755407654, | |
| "grad_norm": 1.9007371664047241, | |
| "learning_rate": 3.613422074320577e-05, | |
| "loss": 0.2516, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.8959426596697811, | |
| "grad_norm": 1.3137174844741821, | |
| "learning_rate": 3.506762233883698e-05, | |
| "loss": 0.2464, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.9599385639319084, | |
| "grad_norm": 1.5280767679214478, | |
| "learning_rate": 3.400102393446819e-05, | |
| "loss": 0.2436, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.2296382337808609, | |
| "eval_mse": 0.22963822030546574, | |
| "eval_runtime": 103.0849, | |
| "eval_samples_per_second": 1766.611, | |
| "eval_steps_per_second": 220.828, | |
| "step": 7813 | |
| }, | |
| { | |
| "epoch": 1.0239344681940357, | |
| "grad_norm": 1.2579801082611084, | |
| "learning_rate": 3.293442553009941e-05, | |
| "loss": 0.2307, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.0879303724561629, | |
| "grad_norm": 1.494954228401184, | |
| "learning_rate": 3.1867827125730624e-05, | |
| "loss": 0.2053, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.15192627671829, | |
| "grad_norm": 1.6762491464614868, | |
| "learning_rate": 3.0801228721361836e-05, | |
| "loss": 0.2021, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.2159221809804173, | |
| "grad_norm": 1.3708257675170898, | |
| "learning_rate": 2.9734630316993045e-05, | |
| "loss": 0.2034, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.2799180852425445, | |
| "grad_norm": 1.57413649559021, | |
| "learning_rate": 2.866803191262426e-05, | |
| "loss": 0.2016, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.3439139895046717, | |
| "grad_norm": 1.1981782913208008, | |
| "learning_rate": 2.7601433508255476e-05, | |
| "loss": 0.2014, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.4079098937667989, | |
| "grad_norm": 2.0707337856292725, | |
| "learning_rate": 2.6534835103886685e-05, | |
| "loss": 0.1996, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.471905798028926, | |
| "grad_norm": 1.3037790060043335, | |
| "learning_rate": 2.54682366995179e-05, | |
| "loss": 0.1998, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.5359017022910533, | |
| "grad_norm": 1.331900715827942, | |
| "learning_rate": 2.4401638295149112e-05, | |
| "loss": 0.1995, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.5998976065531805, | |
| "grad_norm": 1.6664810180664062, | |
| "learning_rate": 2.3335039890780325e-05, | |
| "loss": 0.1995, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.6638935108153077, | |
| "grad_norm": 1.518689751625061, | |
| "learning_rate": 2.2268441486411537e-05, | |
| "loss": 0.1981, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.727889415077435, | |
| "grad_norm": 1.4883790016174316, | |
| "learning_rate": 2.120184308204275e-05, | |
| "loss": 0.1954, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.7918853193395623, | |
| "grad_norm": 1.3662298917770386, | |
| "learning_rate": 2.0135244677673965e-05, | |
| "loss": 0.1957, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.8558812236016895, | |
| "grad_norm": 1.4358400106430054, | |
| "learning_rate": 1.9068646273305177e-05, | |
| "loss": 0.1943, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.9198771278638167, | |
| "grad_norm": 1.3663334846496582, | |
| "learning_rate": 1.800204786893639e-05, | |
| "loss": 0.1959, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.983873032125944, | |
| "grad_norm": 1.3956917524337769, | |
| "learning_rate": 1.69354494645676e-05, | |
| "loss": 0.1927, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.2078620195388794, | |
| "eval_mse": 0.20786202507040266, | |
| "eval_runtime": 99.6454, | |
| "eval_samples_per_second": 1827.591, | |
| "eval_steps_per_second": 228.45, | |
| "step": 15626 | |
| }, | |
| { | |
| "epoch": 2.0478689363880713, | |
| "grad_norm": 1.112040638923645, | |
| "learning_rate": 1.5868851060198814e-05, | |
| "loss": 0.1722, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 2.1118648406501985, | |
| "grad_norm": 1.627584457397461, | |
| "learning_rate": 1.480225265583003e-05, | |
| "loss": 0.1648, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 2.1758607449123257, | |
| "grad_norm": 1.6189327239990234, | |
| "learning_rate": 1.3735654251461241e-05, | |
| "loss": 0.1637, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 2.239856649174453, | |
| "grad_norm": 1.2161808013916016, | |
| "learning_rate": 1.2669055847092454e-05, | |
| "loss": 0.1644, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 2.30385255343658, | |
| "grad_norm": 1.4423365592956543, | |
| "learning_rate": 1.1602457442723666e-05, | |
| "loss": 0.163, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 2.3678484576987073, | |
| "grad_norm": 1.298816204071045, | |
| "learning_rate": 1.053585903835488e-05, | |
| "loss": 0.1626, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 2.4318443619608345, | |
| "grad_norm": 1.2976384162902832, | |
| "learning_rate": 9.469260633986092e-06, | |
| "loss": 0.1627, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 2.4958402662229617, | |
| "grad_norm": 1.39438796043396, | |
| "learning_rate": 8.402662229617304e-06, | |
| "loss": 0.1633, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 2.559836170485089, | |
| "grad_norm": 1.342234492301941, | |
| "learning_rate": 7.336063825248518e-06, | |
| "loss": 0.162, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 2.623832074747216, | |
| "grad_norm": 1.3450177907943726, | |
| "learning_rate": 6.26946542087973e-06, | |
| "loss": 0.1619, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 2.6878279790093433, | |
| "grad_norm": 1.3149842023849487, | |
| "learning_rate": 5.202867016510943e-06, | |
| "loss": 0.1608, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 2.7518238832714705, | |
| "grad_norm": 1.3171745538711548, | |
| "learning_rate": 4.1362686121421564e-06, | |
| "loss": 0.1622, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 2.8158197875335977, | |
| "grad_norm": 1.0992498397827148, | |
| "learning_rate": 3.069670207773369e-06, | |
| "loss": 0.1608, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 2.879815691795725, | |
| "grad_norm": 1.3956327438354492, | |
| "learning_rate": 2.003071803404582e-06, | |
| "loss": 0.161, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 2.943811596057852, | |
| "grad_norm": 1.531733512878418, | |
| "learning_rate": 9.364733990357951e-07, | |
| "loss": 0.1615, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.19583497941493988, | |
| "eval_mse": 0.19583499065839644, | |
| "eval_runtime": 99.9929, | |
| "eval_samples_per_second": 1821.239, | |
| "eval_steps_per_second": 227.656, | |
| "step": 23439 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 23439, | |
| "total_flos": 4.9403660544e+16, | |
| "train_loss": 0.21201796470442558, | |
| "train_runtime": 3304.0326, | |
| "train_samples_per_second": 907.981, | |
| "train_steps_per_second": 7.094 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 23439, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "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": 4.9403660544e+16, | |
| "train_batch_size": 128, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |