| { |
| "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 |
| } |
|
|
