| { | |
| "best_metric": 0.8567830467257253, | |
| "best_model_checkpoint": "/scratch/mrahma45/pixel/finetuned_models/canine/canine-base-finetuned-pos-ud-Vietnamese-VTB/checkpoint-5500", | |
| "epoch": 181.8181818181818, | |
| "global_step": 8000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 8e-05, | |
| "loss": 1.7158, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 4.55, | |
| "learning_rate": 7.946308724832215e-05, | |
| "loss": 0.6087, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 6.82, | |
| "learning_rate": 7.89261744966443e-05, | |
| "loss": 0.3469, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "learning_rate": 7.838926174496645e-05, | |
| "loss": 0.2155, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 11.36, | |
| "learning_rate": 7.78523489932886e-05, | |
| "loss": 0.1352, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 11.36, | |
| "eval_accuracy": 0.845839847142609, | |
| "eval_loss": 0.7401469945907593, | |
| "eval_runtime": 3.1919, | |
| "eval_samples_per_second": 250.631, | |
| "eval_steps_per_second": 31.329, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 13.64, | |
| "learning_rate": 7.731543624161075e-05, | |
| "loss": 0.0882, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 15.91, | |
| "learning_rate": 7.677852348993288e-05, | |
| "loss": 0.0645, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 18.18, | |
| "learning_rate": 7.624161073825503e-05, | |
| "loss": 0.0481, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 20.45, | |
| "learning_rate": 7.570469798657718e-05, | |
| "loss": 0.0386, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 22.73, | |
| "learning_rate": 7.516778523489933e-05, | |
| "loss": 0.0322, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 22.73, | |
| "eval_accuracy": 0.8473163105784263, | |
| "eval_loss": 0.9984953999519348, | |
| "eval_runtime": 3.1616, | |
| "eval_samples_per_second": 253.035, | |
| "eval_steps_per_second": 31.629, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "learning_rate": 7.463087248322148e-05, | |
| "loss": 0.0243, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 27.27, | |
| "learning_rate": 7.409395973154362e-05, | |
| "loss": 0.0221, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 29.55, | |
| "learning_rate": 7.355704697986577e-05, | |
| "loss": 0.0192, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 31.82, | |
| "learning_rate": 7.302013422818792e-05, | |
| "loss": 0.0158, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 34.09, | |
| "learning_rate": 7.248322147651007e-05, | |
| "loss": 0.0182, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 34.09, | |
| "eval_accuracy": 0.8484453708528747, | |
| "eval_loss": 1.106465220451355, | |
| "eval_runtime": 3.1709, | |
| "eval_samples_per_second": 252.294, | |
| "eval_steps_per_second": 31.537, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 36.36, | |
| "learning_rate": 7.194630872483222e-05, | |
| "loss": 0.0173, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 38.64, | |
| "learning_rate": 7.140939597315438e-05, | |
| "loss": 0.0136, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 40.91, | |
| "learning_rate": 7.087248322147653e-05, | |
| "loss": 0.0116, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 43.18, | |
| "learning_rate": 7.033557046979866e-05, | |
| "loss": 0.01, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 45.45, | |
| "learning_rate": 6.979865771812081e-05, | |
| "loss": 0.0112, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 45.45, | |
| "eval_accuracy": 0.8446239360778183, | |
| "eval_loss": 1.2097970247268677, | |
| "eval_runtime": 3.165, | |
| "eval_samples_per_second": 252.768, | |
| "eval_steps_per_second": 31.596, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 47.73, | |
| "learning_rate": 6.926174496644296e-05, | |
| "loss": 0.009, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "learning_rate": 6.87248322147651e-05, | |
| "loss": 0.0103, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 52.27, | |
| "learning_rate": 6.818791946308725e-05, | |
| "loss": 0.0096, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 54.55, | |
| "learning_rate": 6.76510067114094e-05, | |
| "loss": 0.008, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 56.82, | |
| "learning_rate": 6.711409395973155e-05, | |
| "loss": 0.008, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 56.82, | |
| "eval_accuracy": 0.8536564182734063, | |
| "eval_loss": 1.282847285270691, | |
| "eval_runtime": 3.1784, | |
| "eval_samples_per_second": 251.698, | |
| "eval_steps_per_second": 31.462, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 59.09, | |
| "learning_rate": 6.65771812080537e-05, | |
| "loss": 0.0073, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 61.36, | |
| "learning_rate": 6.604026845637585e-05, | |
| "loss": 0.008, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 63.64, | |
| "learning_rate": 6.5503355704698e-05, | |
| "loss": 0.0076, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 65.91, | |
| "learning_rate": 6.496644295302014e-05, | |
| "loss": 0.0072, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 68.18, | |
| "learning_rate": 6.442953020134228e-05, | |
| "loss": 0.0064, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 68.18, | |
| "eval_accuracy": 0.8514851485148515, | |
| "eval_loss": 1.265517234802246, | |
| "eval_runtime": 3.1864, | |
| "eval_samples_per_second": 251.068, | |
| "eval_steps_per_second": 31.384, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 70.45, | |
| "learning_rate": 6.389261744966443e-05, | |
| "loss": 0.0054, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 72.73, | |
| "learning_rate": 6.335570469798657e-05, | |
| "loss": 0.0047, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "learning_rate": 6.281879194630872e-05, | |
| "loss": 0.0064, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 77.27, | |
| "learning_rate": 6.228187919463087e-05, | |
| "loss": 0.0057, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 79.55, | |
| "learning_rate": 6.174496644295302e-05, | |
| "loss": 0.0064, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 79.55, | |
| "eval_accuracy": 0.8527879103699844, | |
| "eval_loss": 1.328222393989563, | |
| "eval_runtime": 3.1799, | |
| "eval_samples_per_second": 251.578, | |
| "eval_steps_per_second": 31.447, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 81.82, | |
| "learning_rate": 6.120805369127517e-05, | |
| "loss": 0.0055, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 84.09, | |
| "learning_rate": 6.067114093959732e-05, | |
| "loss": 0.0056, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 86.36, | |
| "learning_rate": 6.013422818791947e-05, | |
| "loss": 0.0049, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 88.64, | |
| "learning_rate": 5.959731543624162e-05, | |
| "loss": 0.0035, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 90.91, | |
| "learning_rate": 5.906040268456377e-05, | |
| "loss": 0.0045, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 90.91, | |
| "eval_accuracy": 0.8500955358693764, | |
| "eval_loss": 1.3510583639144897, | |
| "eval_runtime": 3.1875, | |
| "eval_samples_per_second": 250.981, | |
| "eval_steps_per_second": 31.373, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 93.18, | |
| "learning_rate": 5.8523489932885916e-05, | |
| "loss": 0.0038, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 95.45, | |
| "learning_rate": 5.798657718120806e-05, | |
| "loss": 0.0033, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 97.73, | |
| "learning_rate": 5.7449664429530206e-05, | |
| "loss": 0.004, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "learning_rate": 5.6912751677852354e-05, | |
| "loss": 0.0042, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 102.27, | |
| "learning_rate": 5.63758389261745e-05, | |
| "loss": 0.0047, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 102.27, | |
| "eval_accuracy": 0.8551328817092235, | |
| "eval_loss": 1.3701601028442383, | |
| "eval_runtime": 3.1715, | |
| "eval_samples_per_second": 252.247, | |
| "eval_steps_per_second": 31.531, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 104.55, | |
| "learning_rate": 5.583892617449665e-05, | |
| "loss": 0.004, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 106.82, | |
| "learning_rate": 5.53020134228188e-05, | |
| "loss": 0.0031, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 109.09, | |
| "learning_rate": 5.476510067114095e-05, | |
| "loss": 0.0047, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 111.36, | |
| "learning_rate": 5.4228187919463095e-05, | |
| "loss": 0.0028, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 113.64, | |
| "learning_rate": 5.369127516778524e-05, | |
| "loss": 0.0039, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 113.64, | |
| "eval_accuracy": 0.8559145388223033, | |
| "eval_loss": 1.4116061925888062, | |
| "eval_runtime": 3.1845, | |
| "eval_samples_per_second": 251.218, | |
| "eval_steps_per_second": 31.402, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 115.91, | |
| "learning_rate": 5.315436241610739e-05, | |
| "loss": 0.004, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 118.18, | |
| "learning_rate": 5.261744966442954e-05, | |
| "loss": 0.0034, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 120.45, | |
| "learning_rate": 5.208053691275168e-05, | |
| "loss": 0.0027, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 122.73, | |
| "learning_rate": 5.154362416107383e-05, | |
| "loss": 0.0023, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 125.0, | |
| "learning_rate": 5.100671140939598e-05, | |
| "loss": 0.0029, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 125.0, | |
| "eval_accuracy": 0.8567830467257253, | |
| "eval_loss": 1.434157371520996, | |
| "eval_runtime": 3.1462, | |
| "eval_samples_per_second": 254.277, | |
| "eval_steps_per_second": 31.785, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 127.27, | |
| "learning_rate": 5.0469798657718126e-05, | |
| "loss": 0.0027, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 129.55, | |
| "learning_rate": 4.9932885906040274e-05, | |
| "loss": 0.0033, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 131.82, | |
| "learning_rate": 4.939597315436242e-05, | |
| "loss": 0.0029, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 134.09, | |
| "learning_rate": 4.885906040268457e-05, | |
| "loss": 0.0023, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 136.36, | |
| "learning_rate": 4.832214765100672e-05, | |
| "loss": 0.0033, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 136.36, | |
| "eval_accuracy": 0.854003821434775, | |
| "eval_loss": 1.4268622398376465, | |
| "eval_runtime": 3.1765, | |
| "eval_samples_per_second": 251.849, | |
| "eval_steps_per_second": 31.481, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 138.64, | |
| "learning_rate": 4.779060402684564e-05, | |
| "loss": 0.0031, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 140.91, | |
| "learning_rate": 4.725369127516779e-05, | |
| "loss": 0.0031, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 143.18, | |
| "learning_rate": 4.6716778523489936e-05, | |
| "loss": 0.0031, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 145.45, | |
| "learning_rate": 4.6179865771812084e-05, | |
| "loss": 0.0024, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 147.73, | |
| "learning_rate": 4.564295302013423e-05, | |
| "loss": 0.0033, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 147.73, | |
| "eval_accuracy": 0.853048462741011, | |
| "eval_loss": 1.4698809385299683, | |
| "eval_runtime": 3.1835, | |
| "eval_samples_per_second": 251.299, | |
| "eval_steps_per_second": 31.412, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 150.0, | |
| "learning_rate": 4.510604026845638e-05, | |
| "loss": 0.0023, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 152.27, | |
| "learning_rate": 4.456912751677853e-05, | |
| "loss": 0.0022, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 154.55, | |
| "learning_rate": 4.403221476510068e-05, | |
| "loss": 0.0026, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 156.82, | |
| "learning_rate": 4.3495302013422825e-05, | |
| "loss": 0.002, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 159.09, | |
| "learning_rate": 4.295838926174497e-05, | |
| "loss": 0.0019, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 159.09, | |
| "eval_accuracy": 0.8554802848705924, | |
| "eval_loss": 1.4916952848434448, | |
| "eval_runtime": 3.1771, | |
| "eval_samples_per_second": 251.799, | |
| "eval_steps_per_second": 31.475, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 161.36, | |
| "learning_rate": 4.242147651006712e-05, | |
| "loss": 0.0032, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 163.64, | |
| "learning_rate": 4.188456375838927e-05, | |
| "loss": 0.002, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 165.91, | |
| "learning_rate": 4.134765100671141e-05, | |
| "loss": 0.0022, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 168.18, | |
| "learning_rate": 4.081073825503356e-05, | |
| "loss": 0.0016, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 170.45, | |
| "learning_rate": 4.027382550335571e-05, | |
| "loss": 0.0018, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 170.45, | |
| "eval_accuracy": 0.854872329338197, | |
| "eval_loss": 1.4624574184417725, | |
| "eval_runtime": 3.1705, | |
| "eval_samples_per_second": 252.327, | |
| "eval_steps_per_second": 31.541, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 172.73, | |
| "learning_rate": 3.9736912751677856e-05, | |
| "loss": 0.0017, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 175.0, | |
| "learning_rate": 3.9200000000000004e-05, | |
| "loss": 0.0014, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 177.27, | |
| "learning_rate": 3.866308724832215e-05, | |
| "loss": 0.0009, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 179.55, | |
| "learning_rate": 3.81261744966443e-05, | |
| "loss": 0.0014, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 181.82, | |
| "learning_rate": 3.758926174496645e-05, | |
| "loss": 0.001, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 181.82, | |
| "eval_accuracy": 0.8567830467257253, | |
| "eval_loss": 1.5747320652008057, | |
| "eval_runtime": 3.1536, | |
| "eval_samples_per_second": 253.674, | |
| "eval_steps_per_second": 31.709, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 181.82, | |
| "step": 8000, | |
| "total_flos": 4.180403182626816e+16, | |
| "train_loss": 0.04628073706757277, | |
| "train_runtime": 1159.4207, | |
| "train_samples_per_second": 414.0, | |
| "train_steps_per_second": 12.937 | |
| } | |
| ], | |
| "max_steps": 15000, | |
| "num_train_epochs": 341, | |
| "total_flos": 4.180403182626816e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |