| { | |
| "best_metric": 0.7404754757881165, | |
| "best_model_checkpoint": "./twitter-xlmr-clip-finetuned-all-123/checkpoint-11500", | |
| "epoch": 1.3828763828763828, | |
| "eval_steps": 500, | |
| "global_step": 11500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 13.600483894348145, | |
| "learning_rate": 4.994047619047619e-05, | |
| "loss": 0.6444, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "eval_f1": 0.41968345507991583, | |
| "eval_loss": 0.8771085143089294, | |
| "eval_precision": 0.6905442329947281, | |
| "eval_recall": 0.4537466834241028, | |
| "eval_runtime": 9.714, | |
| "eval_samples_per_second": 90.385, | |
| "eval_steps_per_second": 5.662, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 3.1294376850128174, | |
| "learning_rate": 4.988035113035113e-05, | |
| "loss": 0.5499, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "eval_f1": 0.41168390354058104, | |
| "eval_loss": 0.8166923522949219, | |
| "eval_precision": 0.7197164083396941, | |
| "eval_recall": 0.4260270911883815, | |
| "eval_runtime": 10.112, | |
| "eval_samples_per_second": 86.827, | |
| "eval_steps_per_second": 5.439, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 3.979219913482666, | |
| "learning_rate": 4.982022607022607e-05, | |
| "loss": 0.5357, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "eval_f1": 0.44236857569210003, | |
| "eval_loss": 0.8084450364112854, | |
| "eval_precision": 0.7263357669536159, | |
| "eval_recall": 0.4695992179863147, | |
| "eval_runtime": 9.4795, | |
| "eval_samples_per_second": 92.621, | |
| "eval_steps_per_second": 5.802, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 4.673430442810059, | |
| "learning_rate": 4.9760341510341516e-05, | |
| "loss": 0.5175, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_f1": 0.37174117930428513, | |
| "eval_loss": 0.8703882098197937, | |
| "eval_precision": 0.6665612592670672, | |
| "eval_recall": 0.426614532420984, | |
| "eval_runtime": 10.0742, | |
| "eval_samples_per_second": 87.153, | |
| "eval_steps_per_second": 5.459, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 4.538591384887695, | |
| "learning_rate": 4.9700216450216456e-05, | |
| "loss": 0.5285, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_f1": 0.42208412407033724, | |
| "eval_loss": 0.906663179397583, | |
| "eval_precision": 0.7528800535330206, | |
| "eval_recall": 0.45646604291765575, | |
| "eval_runtime": 11.1647, | |
| "eval_samples_per_second": 78.641, | |
| "eval_steps_per_second": 4.926, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 3.0253360271453857, | |
| "learning_rate": 4.964009139009139e-05, | |
| "loss": 0.5081, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_f1": 0.6355885803741356, | |
| "eval_loss": 0.7414107322692871, | |
| "eval_precision": 0.7654892081053667, | |
| "eval_recall": 0.6113652655588139, | |
| "eval_runtime": 10.1707, | |
| "eval_samples_per_second": 86.326, | |
| "eval_steps_per_second": 5.408, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 7.686459541320801, | |
| "learning_rate": 4.957996632996633e-05, | |
| "loss": 0.506, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "eval_f1": 0.5785601517423107, | |
| "eval_loss": 0.8712885975837708, | |
| "eval_precision": 0.5829865278619871, | |
| "eval_recall": 0.659145370758274, | |
| "eval_runtime": 9.4937, | |
| "eval_samples_per_second": 92.483, | |
| "eval_steps_per_second": 5.793, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 5.288177967071533, | |
| "learning_rate": 4.951984126984127e-05, | |
| "loss": 0.5049, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_f1": 0.4464146254881079, | |
| "eval_loss": 0.751436173915863, | |
| "eval_precision": 0.5550960735171261, | |
| "eval_recall": 0.4567546432062561, | |
| "eval_runtime": 10.1399, | |
| "eval_samples_per_second": 86.589, | |
| "eval_steps_per_second": 5.424, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 2.32623028755188, | |
| "learning_rate": 4.945971620971621e-05, | |
| "loss": 0.4999, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "eval_f1": 0.5767450661581325, | |
| "eval_loss": 0.7584463357925415, | |
| "eval_precision": 0.6519354622940794, | |
| "eval_recall": 0.5502052785923753, | |
| "eval_runtime": 9.964, | |
| "eval_samples_per_second": 88.117, | |
| "eval_steps_per_second": 5.52, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 8.978556632995605, | |
| "learning_rate": 4.939959114959115e-05, | |
| "loss": 0.507, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_f1": 0.5635568884134308, | |
| "eval_loss": 0.8071982860565186, | |
| "eval_precision": 0.6478587943382395, | |
| "eval_recall": 0.5625815761299632, | |
| "eval_runtime": 9.922, | |
| "eval_samples_per_second": 88.49, | |
| "eval_steps_per_second": 5.543, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 5.773237228393555, | |
| "learning_rate": 4.933946608946609e-05, | |
| "loss": 0.5048, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_f1": 0.5730277716509274, | |
| "eval_loss": 0.8080196380615234, | |
| "eval_precision": 0.6259898141973613, | |
| "eval_recall": 0.5725429409300378, | |
| "eval_runtime": 10.1691, | |
| "eval_samples_per_second": 86.34, | |
| "eval_steps_per_second": 5.409, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 4.596500873565674, | |
| "learning_rate": 4.9279461279461284e-05, | |
| "loss": 0.4907, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "eval_f1": 0.522356162116029, | |
| "eval_loss": 0.7966476082801819, | |
| "eval_precision": 0.6975785731883293, | |
| "eval_recall": 0.5138146441372248, | |
| "eval_runtime": 10.1124, | |
| "eval_samples_per_second": 86.824, | |
| "eval_steps_per_second": 5.439, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 4.815864562988281, | |
| "learning_rate": 4.9219336219336224e-05, | |
| "loss": 0.493, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "eval_f1": 0.4921730387816156, | |
| "eval_loss": 0.8192508220672607, | |
| "eval_precision": 0.7098761086589401, | |
| "eval_recall": 0.4948889819857562, | |
| "eval_runtime": 10.12, | |
| "eval_samples_per_second": 86.759, | |
| "eval_steps_per_second": 5.435, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 2.7458622455596924, | |
| "learning_rate": 4.915921115921116e-05, | |
| "loss": 0.4668, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "eval_f1": 0.6501053053856183, | |
| "eval_loss": 0.7501620650291443, | |
| "eval_precision": 0.6281644378579914, | |
| "eval_recall": 0.6941954103244425, | |
| "eval_runtime": 9.7011, | |
| "eval_samples_per_second": 90.505, | |
| "eval_steps_per_second": 5.669, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 6.319346904754639, | |
| "learning_rate": 4.90990860990861e-05, | |
| "loss": 0.4717, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "eval_f1": 0.5190561121359437, | |
| "eval_loss": 0.7636385560035706, | |
| "eval_precision": 0.6372459240106298, | |
| "eval_recall": 0.5109072289717451, | |
| "eval_runtime": 9.9019, | |
| "eval_samples_per_second": 88.67, | |
| "eval_steps_per_second": 5.554, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 4.232970714569092, | |
| "learning_rate": 4.903896103896104e-05, | |
| "loss": 0.4774, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eval_f1": 0.5587125288791309, | |
| "eval_loss": 0.765232503414154, | |
| "eval_precision": 0.7513256527294695, | |
| "eval_recall": 0.5360405902341386, | |
| "eval_runtime": 10.3268, | |
| "eval_samples_per_second": 85.022, | |
| "eval_steps_per_second": 5.326, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "grad_norm": 6.469419479370117, | |
| "learning_rate": 4.897883597883598e-05, | |
| "loss": 0.4676, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_f1": 0.5836074412050168, | |
| "eval_loss": 0.8481860756874084, | |
| "eval_precision": 0.6371603008082242, | |
| "eval_recall": 0.5918316808639389, | |
| "eval_runtime": 10.1025, | |
| "eval_samples_per_second": 86.909, | |
| "eval_steps_per_second": 5.444, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "grad_norm": 6.76945161819458, | |
| "learning_rate": 4.891871091871092e-05, | |
| "loss": 0.4361, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "eval_f1": 0.5175478304159831, | |
| "eval_loss": 0.7456216812133789, | |
| "eval_precision": 0.668652900688299, | |
| "eval_recall": 0.5176800260671229, | |
| "eval_runtime": 9.5682, | |
| "eval_samples_per_second": 91.762, | |
| "eval_steps_per_second": 5.748, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "grad_norm": 11.62441349029541, | |
| "learning_rate": 4.885858585858586e-05, | |
| "loss": 0.4536, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "eval_f1": 0.5155711517804843, | |
| "eval_loss": 0.8448612689971924, | |
| "eval_precision": 0.73633642018397, | |
| "eval_recall": 0.5160256947353722, | |
| "eval_runtime": 10.0725, | |
| "eval_samples_per_second": 87.168, | |
| "eval_steps_per_second": 5.46, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 6.764497756958008, | |
| "learning_rate": 4.879858104858105e-05, | |
| "loss": 0.4277, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "eval_f1": 0.5173349712389866, | |
| "eval_loss": 0.8647661805152893, | |
| "eval_precision": 0.6381631231821719, | |
| "eval_recall": 0.5247414234511009, | |
| "eval_runtime": 10.0414, | |
| "eval_samples_per_second": 87.438, | |
| "eval_steps_per_second": 5.477, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "grad_norm": 6.405815601348877, | |
| "learning_rate": 4.873845598845599e-05, | |
| "loss": 0.4444, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "eval_f1": 0.5958880582085585, | |
| "eval_loss": 0.8722940683364868, | |
| "eval_precision": 0.5870961573122971, | |
| "eval_recall": 0.6621933621933621, | |
| "eval_runtime": 10.0728, | |
| "eval_samples_per_second": 87.166, | |
| "eval_steps_per_second": 5.46, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "grad_norm": 3.5875844955444336, | |
| "learning_rate": 4.8678330928330925e-05, | |
| "loss": 0.4269, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "eval_f1": 0.5525570018606619, | |
| "eval_loss": 0.7856002449989319, | |
| "eval_precision": 0.6150631452676062, | |
| "eval_recall": 0.5520914211236793, | |
| "eval_runtime": 9.6968, | |
| "eval_samples_per_second": 90.545, | |
| "eval_steps_per_second": 5.672, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "grad_norm": 8.949612617492676, | |
| "learning_rate": 4.8618205868205866e-05, | |
| "loss": 0.4322, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "eval_f1": 0.6400986522737416, | |
| "eval_loss": 0.7404754757881165, | |
| "eval_precision": 0.6430963017398649, | |
| "eval_recall": 0.6553870502257598, | |
| "eval_runtime": 10.249, | |
| "eval_samples_per_second": 85.667, | |
| "eval_steps_per_second": 5.366, | |
| "step": 11500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 415800, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "total_flos": 0.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |