| { | |
| "best_metric": 0.4452793300151825, | |
| "best_model_checkpoint": "./models/adapters_mlm_cn/bg/checkpoint-36000", | |
| "epoch": 11.022657685241887, | |
| "eval_steps": 500, | |
| "global_step": 36000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 4.9500000000000004e-05, | |
| "loss": 1.5057, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_accuracy": 0.8148620791204476, | |
| "eval_loss": 0.9846400618553162, | |
| "eval_runtime": 7.7173, | |
| "eval_samples_per_second": 752.339, | |
| "eval_steps_per_second": 47.037, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 4.9e-05, | |
| "loss": 1.0172, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eval_accuracy": 0.82587890625, | |
| "eval_loss": 0.8394753932952881, | |
| "eval_runtime": 7.7319, | |
| "eval_samples_per_second": 750.918, | |
| "eval_steps_per_second": 46.949, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 4.85e-05, | |
| "loss": 0.8814, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "eval_accuracy": 0.8368038740920097, | |
| "eval_loss": 0.7822620272636414, | |
| "eval_runtime": 7.7294, | |
| "eval_samples_per_second": 751.162, | |
| "eval_steps_per_second": 46.964, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.8405, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "eval_accuracy": 0.8449071800412533, | |
| "eval_loss": 0.7436666488647461, | |
| "eval_runtime": 7.9259, | |
| "eval_samples_per_second": 732.538, | |
| "eval_steps_per_second": 45.799, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.7773, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "eval_accuracy": 0.8386841062227507, | |
| "eval_loss": 0.7246997952461243, | |
| "eval_runtime": 7.7331, | |
| "eval_samples_per_second": 750.803, | |
| "eval_steps_per_second": 46.941, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.7762, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "eval_accuracy": 0.8513044340839202, | |
| "eval_loss": 0.6520901322364807, | |
| "eval_runtime": 7.9369, | |
| "eval_samples_per_second": 731.516, | |
| "eval_steps_per_second": 45.736, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 4.6500000000000005e-05, | |
| "loss": 0.7186, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "eval_accuracy": 0.8492265517916585, | |
| "eval_loss": 0.6834315061569214, | |
| "eval_runtime": 7.6738, | |
| "eval_samples_per_second": 756.596, | |
| "eval_steps_per_second": 47.304, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.7033, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "eval_accuracy": 0.852271607371637, | |
| "eval_loss": 0.67154860496521, | |
| "eval_runtime": 7.7294, | |
| "eval_samples_per_second": 751.163, | |
| "eval_steps_per_second": 46.964, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 4.55e-05, | |
| "loss": 0.672, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "eval_accuracy": 0.855973974763407, | |
| "eval_loss": 0.6539207100868225, | |
| "eval_runtime": 7.7117, | |
| "eval_samples_per_second": 752.881, | |
| "eval_steps_per_second": 47.071, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.6613, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "eval_accuracy": 0.8567085131424088, | |
| "eval_loss": 0.638721227645874, | |
| "eval_runtime": 7.6505, | |
| "eval_samples_per_second": 758.9, | |
| "eval_steps_per_second": 47.448, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 4.4500000000000004e-05, | |
| "loss": 0.6712, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "eval_accuracy": 0.862372613040467, | |
| "eval_loss": 0.6180465221405029, | |
| "eval_runtime": 7.7012, | |
| "eval_samples_per_second": 753.913, | |
| "eval_steps_per_second": 47.136, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.6776, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "eval_accuracy": 0.8537038849202466, | |
| "eval_loss": 0.6634594202041626, | |
| "eval_runtime": 7.7042, | |
| "eval_samples_per_second": 753.61, | |
| "eval_steps_per_second": 47.117, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 4.35e-05, | |
| "loss": 0.6484, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "eval_accuracy": 0.8661394258933802, | |
| "eval_loss": 0.5945894122123718, | |
| "eval_runtime": 7.6974, | |
| "eval_samples_per_second": 754.283, | |
| "eval_steps_per_second": 47.159, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.6817, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "eval_accuracy": 0.8654563297350344, | |
| "eval_loss": 0.6126104593276978, | |
| "eval_runtime": 8.509, | |
| "eval_samples_per_second": 682.334, | |
| "eval_steps_per_second": 42.661, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.6392, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "eval_accuracy": 0.8613216715257531, | |
| "eval_loss": 0.613590657711029, | |
| "eval_runtime": 8.1378, | |
| "eval_samples_per_second": 713.458, | |
| "eval_steps_per_second": 44.606, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.6394, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "eval_accuracy": 0.8620723749258453, | |
| "eval_loss": 0.6320650577545166, | |
| "eval_runtime": 7.7697, | |
| "eval_samples_per_second": 747.26, | |
| "eval_steps_per_second": 46.72, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 4.15e-05, | |
| "loss": 0.6273, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "eval_accuracy": 0.8629402009560043, | |
| "eval_loss": 0.5997043251991272, | |
| "eval_runtime": 7.9947, | |
| "eval_samples_per_second": 726.232, | |
| "eval_steps_per_second": 45.405, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.5993, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "eval_accuracy": 0.8645569620253165, | |
| "eval_loss": 0.6027613282203674, | |
| "eval_runtime": 8.0195, | |
| "eval_samples_per_second": 723.989, | |
| "eval_steps_per_second": 45.265, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 4.05e-05, | |
| "loss": 0.6527, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "eval_accuracy": 0.8510214250124564, | |
| "eval_loss": 0.6583752632141113, | |
| "eval_runtime": 7.9195, | |
| "eval_samples_per_second": 733.131, | |
| "eval_steps_per_second": 45.836, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 4e-05, | |
| "loss": 0.5897, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "eval_accuracy": 0.8676120587068623, | |
| "eval_loss": 0.5727556943893433, | |
| "eval_runtime": 7.9746, | |
| "eval_samples_per_second": 728.065, | |
| "eval_steps_per_second": 45.52, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "learning_rate": 3.9500000000000005e-05, | |
| "loss": 0.574, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "eval_accuracy": 0.8670824400701618, | |
| "eval_loss": 0.5869864821434021, | |
| "eval_runtime": 7.8716, | |
| "eval_samples_per_second": 737.59, | |
| "eval_steps_per_second": 46.115, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "learning_rate": 3.9000000000000006e-05, | |
| "loss": 0.6026, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "eval_accuracy": 0.8676513458361675, | |
| "eval_loss": 0.6066599488258362, | |
| "eval_runtime": 7.8242, | |
| "eval_samples_per_second": 742.057, | |
| "eval_steps_per_second": 46.395, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "learning_rate": 3.85e-05, | |
| "loss": 0.5896, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "eval_accuracy": 0.8638327806250629, | |
| "eval_loss": 0.6000019311904907, | |
| "eval_runtime": 8.0139, | |
| "eval_samples_per_second": 724.49, | |
| "eval_steps_per_second": 45.296, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "learning_rate": 3.8e-05, | |
| "loss": 0.566, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "eval_accuracy": 0.8711821948164563, | |
| "eval_loss": 0.5566375851631165, | |
| "eval_runtime": 7.8868, | |
| "eval_samples_per_second": 736.17, | |
| "eval_steps_per_second": 46.026, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.5928, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "eval_accuracy": 0.8675352877307275, | |
| "eval_loss": 0.5621004700660706, | |
| "eval_runtime": 7.9912, | |
| "eval_samples_per_second": 726.553, | |
| "eval_steps_per_second": 45.425, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "learning_rate": 3.7e-05, | |
| "loss": 0.597, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "eval_accuracy": 0.8771320904403015, | |
| "eval_loss": 0.5161893963813782, | |
| "eval_runtime": 7.9771, | |
| "eval_samples_per_second": 727.833, | |
| "eval_steps_per_second": 45.505, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 4.13, | |
| "learning_rate": 3.65e-05, | |
| "loss": 0.5836, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 4.13, | |
| "eval_accuracy": 0.8696463654223968, | |
| "eval_loss": 0.5498046278953552, | |
| "eval_runtime": 7.8463, | |
| "eval_samples_per_second": 739.966, | |
| "eval_steps_per_second": 46.264, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 4.29, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.5864, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 4.29, | |
| "eval_accuracy": 0.8639773945240183, | |
| "eval_loss": 0.5728442072868347, | |
| "eval_runtime": 7.8404, | |
| "eval_samples_per_second": 740.524, | |
| "eval_steps_per_second": 46.299, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "learning_rate": 3.55e-05, | |
| "loss": 0.5562, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "eval_accuracy": 0.8623497479643273, | |
| "eval_loss": 0.6000498533248901, | |
| "eval_runtime": 7.8135, | |
| "eval_samples_per_second": 743.077, | |
| "eval_steps_per_second": 46.458, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 4.59, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.5999, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 4.59, | |
| "eval_accuracy": 0.8679152291769344, | |
| "eval_loss": 0.5589025020599365, | |
| "eval_runtime": 7.7959, | |
| "eval_samples_per_second": 744.749, | |
| "eval_steps_per_second": 46.563, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 4.75, | |
| "learning_rate": 3.45e-05, | |
| "loss": 0.5767, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 4.75, | |
| "eval_accuracy": 0.8680821783151479, | |
| "eval_loss": 0.5713112354278564, | |
| "eval_runtime": 8.9874, | |
| "eval_samples_per_second": 646.014, | |
| "eval_steps_per_second": 40.39, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "learning_rate": 3.4000000000000007e-05, | |
| "loss": 0.5574, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "eval_accuracy": 0.8739122026687295, | |
| "eval_loss": 0.5337920784950256, | |
| "eval_runtime": 10.8383, | |
| "eval_samples_per_second": 535.691, | |
| "eval_steps_per_second": 33.492, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "learning_rate": 3.35e-05, | |
| "loss": 0.568, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "eval_accuracy": 0.87250098000784, | |
| "eval_loss": 0.552727222442627, | |
| "eval_runtime": 7.8124, | |
| "eval_samples_per_second": 743.18, | |
| "eval_steps_per_second": 46.465, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 5.21, | |
| "learning_rate": 3.3e-05, | |
| "loss": 0.5568, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 5.21, | |
| "eval_accuracy": 0.8776927722971612, | |
| "eval_loss": 0.5058096051216125, | |
| "eval_runtime": 7.8143, | |
| "eval_samples_per_second": 742.993, | |
| "eval_steps_per_second": 46.453, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 5.36, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.5369, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 5.36, | |
| "eval_accuracy": 0.8719769673704415, | |
| "eval_loss": 0.5599194169044495, | |
| "eval_runtime": 7.8287, | |
| "eval_samples_per_second": 741.628, | |
| "eval_steps_per_second": 46.368, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 5.51, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 0.518, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 5.51, | |
| "eval_accuracy": 0.8720388349514563, | |
| "eval_loss": 0.561033308506012, | |
| "eval_runtime": 7.8241, | |
| "eval_samples_per_second": 742.071, | |
| "eval_steps_per_second": 46.395, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 5.66, | |
| "learning_rate": 3.15e-05, | |
| "loss": 0.5637, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 5.66, | |
| "eval_accuracy": 0.8727518855153742, | |
| "eval_loss": 0.5467284917831421, | |
| "eval_runtime": 8.0155, | |
| "eval_samples_per_second": 724.347, | |
| "eval_steps_per_second": 45.287, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 5.82, | |
| "learning_rate": 3.1e-05, | |
| "loss": 0.557, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 5.82, | |
| "eval_accuracy": 0.8713813872158539, | |
| "eval_loss": 0.5348953604698181, | |
| "eval_runtime": 8.0121, | |
| "eval_samples_per_second": 724.653, | |
| "eval_steps_per_second": 45.306, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 5.97, | |
| "learning_rate": 3.05e-05, | |
| "loss": 0.5499, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 5.97, | |
| "eval_accuracy": 0.8724001160878398, | |
| "eval_loss": 0.5467893481254578, | |
| "eval_runtime": 7.7511, | |
| "eval_samples_per_second": 749.05, | |
| "eval_steps_per_second": 46.832, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 6.12, | |
| "learning_rate": 3e-05, | |
| "loss": 0.5304, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 6.12, | |
| "eval_accuracy": 0.8740521910388971, | |
| "eval_loss": 0.5243064761161804, | |
| "eval_runtime": 7.8201, | |
| "eval_samples_per_second": 742.45, | |
| "eval_steps_per_second": 46.419, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 6.28, | |
| "learning_rate": 2.95e-05, | |
| "loss": 0.5431, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 6.28, | |
| "eval_accuracy": 0.8783942176206291, | |
| "eval_loss": 0.4997641146183014, | |
| "eval_runtime": 8.0018, | |
| "eval_samples_per_second": 725.585, | |
| "eval_steps_per_second": 45.365, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 6.43, | |
| "learning_rate": 2.9e-05, | |
| "loss": 0.5508, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 6.43, | |
| "eval_accuracy": 0.8763812154696132, | |
| "eval_loss": 0.5366745591163635, | |
| "eval_runtime": 7.8074, | |
| "eval_samples_per_second": 743.654, | |
| "eval_steps_per_second": 46.494, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 6.58, | |
| "learning_rate": 2.8499999999999998e-05, | |
| "loss": 0.5701, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 6.58, | |
| "eval_accuracy": 0.8734250823803063, | |
| "eval_loss": 0.5364522337913513, | |
| "eval_runtime": 7.9868, | |
| "eval_samples_per_second": 726.947, | |
| "eval_steps_per_second": 45.45, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 6.74, | |
| "learning_rate": 2.8000000000000003e-05, | |
| "loss": 0.521, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 6.74, | |
| "eval_accuracy": 0.8818635607321131, | |
| "eval_loss": 0.4879148006439209, | |
| "eval_runtime": 7.9938, | |
| "eval_samples_per_second": 726.31, | |
| "eval_steps_per_second": 45.41, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 6.89, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.5514, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 6.89, | |
| "eval_accuracy": 0.8786950074147306, | |
| "eval_loss": 0.5105842351913452, | |
| "eval_runtime": 7.8325, | |
| "eval_samples_per_second": 741.269, | |
| "eval_steps_per_second": 46.345, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "learning_rate": 2.7000000000000002e-05, | |
| "loss": 0.547, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "eval_accuracy": 0.8747058823529412, | |
| "eval_loss": 0.5258113741874695, | |
| "eval_runtime": 7.8237, | |
| "eval_samples_per_second": 742.1, | |
| "eval_steps_per_second": 46.397, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 7.2, | |
| "learning_rate": 2.6500000000000004e-05, | |
| "loss": 0.5512, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 7.2, | |
| "eval_accuracy": 0.877830692973078, | |
| "eval_loss": 0.49750423431396484, | |
| "eval_runtime": 7.9086, | |
| "eval_samples_per_second": 734.135, | |
| "eval_steps_per_second": 45.899, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 7.35, | |
| "learning_rate": 2.6000000000000002e-05, | |
| "loss": 0.5407, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 7.35, | |
| "eval_accuracy": 0.8785601265822784, | |
| "eval_loss": 0.494391530752182, | |
| "eval_runtime": 8.2168, | |
| "eval_samples_per_second": 706.599, | |
| "eval_steps_per_second": 44.178, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "learning_rate": 2.5500000000000003e-05, | |
| "loss": 0.5181, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "eval_accuracy": 0.8794734275962945, | |
| "eval_loss": 0.4911736845970154, | |
| "eval_runtime": 8.2044, | |
| "eval_samples_per_second": 707.673, | |
| "eval_steps_per_second": 44.245, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 7.65, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.5493, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 7.65, | |
| "eval_accuracy": 0.87302207462395, | |
| "eval_loss": 0.5187950730323792, | |
| "eval_runtime": 8.0486, | |
| "eval_samples_per_second": 721.366, | |
| "eval_steps_per_second": 45.101, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 7.81, | |
| "learning_rate": 2.45e-05, | |
| "loss": 0.5388, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 7.81, | |
| "eval_accuracy": 0.8831105473704751, | |
| "eval_loss": 0.5000073313713074, | |
| "eval_runtime": 8.0362, | |
| "eval_samples_per_second": 722.481, | |
| "eval_steps_per_second": 45.171, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 7.96, | |
| "learning_rate": 2.4e-05, | |
| "loss": 0.5284, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 7.96, | |
| "eval_accuracy": 0.8737309019221291, | |
| "eval_loss": 0.5161213278770447, | |
| "eval_runtime": 8.1271, | |
| "eval_samples_per_second": 714.401, | |
| "eval_steps_per_second": 44.665, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 8.11, | |
| "learning_rate": 2.35e-05, | |
| "loss": 0.5116, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 8.11, | |
| "eval_accuracy": 0.8759842519685039, | |
| "eval_loss": 0.5262829065322876, | |
| "eval_runtime": 8.1731, | |
| "eval_samples_per_second": 710.381, | |
| "eval_steps_per_second": 44.414, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 8.27, | |
| "learning_rate": 2.3000000000000003e-05, | |
| "loss": 0.5161, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 8.27, | |
| "eval_accuracy": 0.8786888397577621, | |
| "eval_loss": 0.500228762626648, | |
| "eval_runtime": 8.034, | |
| "eval_samples_per_second": 722.681, | |
| "eval_steps_per_second": 45.183, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 8.42, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.5185, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 8.42, | |
| "eval_accuracy": 0.8744550138723741, | |
| "eval_loss": 0.5127227902412415, | |
| "eval_runtime": 7.7182, | |
| "eval_samples_per_second": 752.252, | |
| "eval_steps_per_second": 47.032, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 8.57, | |
| "learning_rate": 2.2000000000000003e-05, | |
| "loss": 0.5291, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 8.57, | |
| "eval_accuracy": 0.8782496527088708, | |
| "eval_loss": 0.5115563273429871, | |
| "eval_runtime": 8.0802, | |
| "eval_samples_per_second": 718.543, | |
| "eval_steps_per_second": 44.924, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 8.73, | |
| "learning_rate": 2.15e-05, | |
| "loss": 0.5061, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 8.73, | |
| "eval_accuracy": 0.8773942634905202, | |
| "eval_loss": 0.4972003996372223, | |
| "eval_runtime": 7.7937, | |
| "eval_samples_per_second": 744.959, | |
| "eval_steps_per_second": 46.576, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 8.88, | |
| "learning_rate": 2.1e-05, | |
| "loss": 0.479, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 8.88, | |
| "eval_accuracy": 0.8797814207650273, | |
| "eval_loss": 0.49780747294425964, | |
| "eval_runtime": 7.8838, | |
| "eval_samples_per_second": 736.449, | |
| "eval_steps_per_second": 46.044, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 2.05e-05, | |
| "loss": 0.5154, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "eval_accuracy": 0.877119904790241, | |
| "eval_loss": 0.5088150501251221, | |
| "eval_runtime": 7.7019, | |
| "eval_samples_per_second": 753.843, | |
| "eval_steps_per_second": 47.131, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 9.19, | |
| "learning_rate": 2e-05, | |
| "loss": 0.4989, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 9.19, | |
| "eval_accuracy": 0.8744332741967278, | |
| "eval_loss": 0.5118668079376221, | |
| "eval_runtime": 7.7316, | |
| "eval_samples_per_second": 750.942, | |
| "eval_steps_per_second": 46.95, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 9.34, | |
| "learning_rate": 1.9500000000000003e-05, | |
| "loss": 0.5098, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 9.34, | |
| "eval_accuracy": 0.8825979176995538, | |
| "eval_loss": 0.4915599524974823, | |
| "eval_runtime": 7.7036, | |
| "eval_samples_per_second": 753.676, | |
| "eval_steps_per_second": 47.121, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 9.49, | |
| "learning_rate": 1.9e-05, | |
| "loss": 0.4777, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 9.49, | |
| "eval_accuracy": 0.882445449184195, | |
| "eval_loss": 0.49568039178848267, | |
| "eval_runtime": 7.7025, | |
| "eval_samples_per_second": 753.779, | |
| "eval_steps_per_second": 47.127, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 9.64, | |
| "learning_rate": 1.85e-05, | |
| "loss": 0.5462, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 9.64, | |
| "eval_accuracy": 0.8778625954198473, | |
| "eval_loss": 0.48457399010658264, | |
| "eval_runtime": 7.7115, | |
| "eval_samples_per_second": 752.903, | |
| "eval_steps_per_second": 47.073, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 9.8, | |
| "learning_rate": 1.8e-05, | |
| "loss": 0.509, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 9.8, | |
| "eval_accuracy": 0.8810146190337884, | |
| "eval_loss": 0.48734790086746216, | |
| "eval_runtime": 7.7302, | |
| "eval_samples_per_second": 751.078, | |
| "eval_steps_per_second": 46.959, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 9.95, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.5181, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 9.95, | |
| "eval_accuracy": 0.8710217755443886, | |
| "eval_loss": 0.5227355360984802, | |
| "eval_runtime": 7.7073, | |
| "eval_samples_per_second": 753.31, | |
| "eval_steps_per_second": 47.098, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 10.1, | |
| "learning_rate": 1.7000000000000003e-05, | |
| "loss": 0.5269, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 10.1, | |
| "eval_accuracy": 0.8802636757182212, | |
| "eval_loss": 0.49287834763526917, | |
| "eval_runtime": 8.3473, | |
| "eval_samples_per_second": 695.551, | |
| "eval_steps_per_second": 43.487, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 10.26, | |
| "learning_rate": 1.65e-05, | |
| "loss": 0.5094, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 10.26, | |
| "eval_accuracy": 0.8877481840193705, | |
| "eval_loss": 0.4840761125087738, | |
| "eval_runtime": 8.4693, | |
| "eval_samples_per_second": 685.535, | |
| "eval_steps_per_second": 42.861, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 10.41, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.5033, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 10.41, | |
| "eval_accuracy": 0.8805490654205608, | |
| "eval_loss": 0.5128547549247742, | |
| "eval_runtime": 8.1006, | |
| "eval_samples_per_second": 716.736, | |
| "eval_steps_per_second": 44.811, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 10.56, | |
| "learning_rate": 1.55e-05, | |
| "loss": 0.4913, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 10.56, | |
| "eval_accuracy": 0.8789432939810334, | |
| "eval_loss": 0.4978225529193878, | |
| "eval_runtime": 8.4845, | |
| "eval_samples_per_second": 684.304, | |
| "eval_steps_per_second": 42.784, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 10.72, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.4938, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 10.72, | |
| "eval_accuracy": 0.8838202465301368, | |
| "eval_loss": 0.46402791142463684, | |
| "eval_runtime": 8.2894, | |
| "eval_samples_per_second": 700.417, | |
| "eval_steps_per_second": 43.791, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 10.87, | |
| "learning_rate": 1.45e-05, | |
| "loss": 0.4954, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 10.87, | |
| "eval_accuracy": 0.8793576184880533, | |
| "eval_loss": 0.4990694522857666, | |
| "eval_runtime": 8.2824, | |
| "eval_samples_per_second": 701.003, | |
| "eval_steps_per_second": 43.828, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 11.02, | |
| "learning_rate": 1.4000000000000001e-05, | |
| "loss": 0.458, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 11.02, | |
| "eval_accuracy": 0.8885711468297012, | |
| "eval_loss": 0.4452793300151825, | |
| "eval_runtime": 8.4212, | |
| "eval_samples_per_second": 689.451, | |
| "eval_steps_per_second": 43.106, | |
| "step": 36000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 50000, | |
| "num_train_epochs": 16, | |
| "save_steps": 500, | |
| "total_flos": 5426535158775808.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |