End of training

README.md

@@ -0,0 +1,115 @@
+---
+license: apache-2.0
+base_model: google/electra-base-generator
+tags:
+- generated_from_trainer
+datasets:
+- datasets/all_binary_and_xe_ey_fae_counterfactual
+metrics:
+- accuracy
+model-index:
+- name: electra-adapter-finetuned-xe_ey_fae
+  results:
+  - task:
+      name: Masked Language Modeling
+      type: fill-mask
+    dataset:
+      name: datasets/all_binary_and_xe_ey_fae_counterfactual
+      type: datasets/all_binary_and_xe_ey_fae_counterfactual
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.6258363412553052
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# electra-adapter-finetuned-xe_ey_fae
+
+This model is a fine-tuned version of [google/electra-base-generator](https://huggingface.co/google/electra-base-generator) on the datasets/all_binary_and_xe_ey_fae_counterfactual dataset.
+It achieves the following results on the evaluation set:
+- Loss: 2.0392
+- Accuracy: 0.6258
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 1e-05
+- train_batch_size: 8
+- eval_batch_size: 8
+- seed: 100
+- gradient_accumulation_steps: 2
+- total_train_batch_size: 16
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- num_epochs: 3.0
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch | Step  | Validation Loss | Accuracy |
+|:-------------:|:-----:|:-----:|:---------------:|:--------:|
+| 3.9488        | 0.06  | 500   | 3.1500          | 0.5509   |
+| 2.942         | 0.13  | 1000  | 2.5844          | 0.5680   |
+| 2.6751        | 0.19  | 1500  | 2.4443          | 0.5790   |
+| 2.582         | 0.26  | 2000  | 2.3701          | 0.5869   |
+| 2.5267        | 0.32  | 2500  | 2.3097          | 0.5937   |
+| 2.4722        | 0.39  | 3000  | 2.2695          | 0.5986   |
+| 2.4289        | 0.45  | 3500  | 2.2329          | 0.6024   |
+| 2.404         | 0.52  | 4000  | 2.2063          | 0.6055   |
+| 2.3826        | 0.58  | 4500  | 2.1840          | 0.6087   |
+| 2.3633        | 0.64  | 5000  | 2.1646          | 0.6109   |
+| 2.3425        | 0.71  | 5500  | 2.1557          | 0.6121   |
+| 2.333         | 0.77  | 6000  | 2.1350          | 0.6141   |
+| 2.311         | 0.84  | 6500  | 2.1292          | 0.6152   |
+| 2.3014        | 0.9   | 7000  | 2.1182          | 0.6166   |
+| 2.2974        | 0.97  | 7500  | 2.1121          | 0.6170   |
+| 2.2866        | 1.03  | 8000  | 2.1079          | 0.6173   |
+| 2.2675        | 1.1   | 8500  | 2.0940          | 0.6192   |
+| 2.2789        | 1.16  | 9000  | 2.0882          | 0.6201   |
+| 2.2684        | 1.22  | 9500  | 2.0873          | 0.6200   |
+| 2.2608        | 1.29  | 10000 | 2.0796          | 0.6209   |
+| 2.2478        | 1.35  | 10500 | 2.0827          | 0.6204   |
+| 2.2524        | 1.42  | 11000 | 2.0741          | 0.6215   |
+| 2.2502        | 1.48  | 11500 | 2.0685          | 0.6220   |
+| 2.243         | 1.55  | 12000 | 2.0665          | 0.6228   |
+| 2.2417        | 1.61  | 12500 | 2.0632          | 0.6229   |
+| 2.2398        | 1.68  | 13000 | 2.0593          | 0.6232   |
+| 2.2233        | 1.74  | 13500 | 2.0600          | 0.6232   |
+| 2.2277        | 1.8   | 14000 | 2.0535          | 0.6236   |
+| 2.2344        | 1.87  | 14500 | 2.0485          | 0.6248   |
+| 2.2274        | 1.93  | 15000 | 2.0507          | 0.6245   |
+| 2.2212        | 2.0   | 15500 | 2.0428          | 0.6256   |
+| 2.214         | 2.06  | 16000 | 2.0464          | 0.6244   |
+| 2.2104        | 2.13  | 16500 | 2.0477          | 0.6250   |
+| 2.2185        | 2.19  | 17000 | 2.0397          | 0.6257   |
+| 2.2157        | 2.26  | 17500 | 2.0419          | 0.6257   |
+| 2.2128        | 2.32  | 18000 | 2.0439          | 0.6255   |
+| 2.2154        | 2.38  | 18500 | 2.0372          | 0.6259   |
+| 2.2099        | 2.45  | 19000 | 2.0337          | 0.6263   |
+| 2.2045        | 2.51  | 19500 | 2.0396          | 0.6259   |
+| 2.2138        | 2.58  | 20000 | 2.0390          | 0.6262   |
+| 2.2103        | 2.64  | 20500 | 2.0339          | 0.6263   |
+
+
+### Framework versions
+
+- Transformers 4.36.2
+- Pytorch 2.2.0+cu121
+- Datasets 2.17.0
+- Tokenizers 0.15.2

all_results.json

@@ -0,0 +1,15 @@
+{
+    "epoch": 2.64,
+    "eval_accuracy": 0.6258363412553052,
+    "eval_loss": 2.0392136573791504,
+    "eval_runtime": 87.4708,
+    "eval_samples": 15525,
+    "eval_samples_per_second": 177.488,
+    "eval_steps_per_second": 22.19,
+    "perplexity": 7.684564122147852,
+    "train_loss": 2.351401915015244,
+    "train_runtime": 7059.8447,
+    "train_samples": 124124,
+    "train_samples_per_second": 52.745,
+    "train_steps_per_second": 3.297
+}

datasets/all_binary_and_xe_ey_fae_counterfactual/adapter_config.json

@@ -0,0 +1,41 @@
+{
+  "config": {
+    "adapter_residual_before_ln": false,
+    "cross_adapter": false,
+    "factorized_phm_W": true,
+    "factorized_phm_rule": false,
+    "hypercomplex_nonlinearity": "glorot-uniform",
+    "init_weights": "bert",
+    "inv_adapter": null,
+    "inv_adapter_reduction_factor": null,
+    "is_parallel": false,
+    "learn_phm": true,
+    "leave_out": [],
+    "ln_after": false,
+    "ln_before": false,
+    "mh_adapter": false,
+    "non_linearity": "relu",
+    "original_ln_after": true,
+    "original_ln_before": true,
+    "output_adapter": true,
+    "phm_bias": true,
+    "phm_c_init": "normal",
+    "phm_dim": 4,
+    "phm_init_range": 0.0001,
+    "phm_layer": false,
+    "phm_rank": 1,
+    "reduction_factor": 16,
+    "residual_before_ln": true,
+    "scaling": 1.0,
+    "shared_W_phm": false,
+    "shared_phm_rule": true,
+    "use_gating": false
+  },
+  "config_id": "9076f36a74755ac4",
+  "hidden_size": 256,
+  "model_class": "ElectraForMaskedLM",
+  "model_name": "google/electra-base-generator",
+  "model_type": "electra",
+  "name": "datasets/all_binary_and_xe_ey_fae_counterfactual",
+  "version": "0.1.2"
+}

datasets/all_binary_and_xe_ey_fae_counterfactual/head_config.json

@@ -0,0 +1,14 @@
+{
+  "config": null,
+  "hidden_size": 256,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1
+  },
+  "model_class": "ElectraForMaskedLM",
+  "model_name": "google/electra-base-generator",
+  "model_type": "electra",
+  "name": null,
+  "num_labels": 2,
+  "version": "0.1.2"
+}

datasets/all_binary_and_xe_ey_fae_counterfactual/pytorch_adapter.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:48344746d047a38e3f7159048f13f2271dd6ce2f8ee3e79f55ed1043d9b44f21
+size 425830

datasets/all_binary_and_xe_ey_fae_counterfactual/pytorch_model_head.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2c0648fbb6aa6e8b4fd9494a9da8a1ff89ca22bc8ff2ead6f16d707f696a993c
+size 94684086

eval_results.json

@@ -0,0 +1,10 @@
+{
+    "epoch": 2.64,
+    "eval_accuracy": 0.6258363412553052,
+    "eval_loss": 2.0392136573791504,
+    "eval_runtime": 87.4708,
+    "eval_samples": 15525,
+    "eval_samples_per_second": 177.488,
+    "eval_steps_per_second": 22.19,
+    "perplexity": 7.684564122147852
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.64,
+    "train_loss": 2.351401915015244,
+    "train_runtime": 7059.8447,
+    "train_samples": 124124,
+    "train_samples_per_second": 52.745,
+    "train_steps_per_second": 3.297
+}

trainer_state.json

@@ -0,0 +1,645 @@
+{
+  "best_metric": 2.0336806774139404,
+  "best_model_checkpoint": "finetuning/output/electra-adapter-finetuned_xe_ey_fae/checkpoint-19000",
+  "epoch": 2.642433616911575,
+  "eval_steps": 500,
+  "global_step": 20500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.06,
+      "learning_rate": 9.785167998625076e-06,
+      "loss": 3.9488,
+      "step": 500
+    },
+    {
+      "epoch": 0.06,
+      "eval_accuracy": 0.5508950432882589,
+      "eval_loss": 3.1499977111816406,
+      "eval_runtime": 85.1217,
+      "eval_samples_per_second": 182.386,
+      "eval_steps_per_second": 22.803,
+      "step": 500
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 9.57033599725015e-06,
+      "loss": 2.942,
+      "step": 1000
+    },
+    {
+      "epoch": 0.13,
+      "eval_accuracy": 0.5680209177510359,
+      "eval_loss": 2.584392547607422,
+      "eval_runtime": 79.4716,
+      "eval_samples_per_second": 195.353,
+      "eval_steps_per_second": 24.424,
+      "step": 1000
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 9.355503995875225e-06,
+      "loss": 2.6751,
+      "step": 1500
+    },
+    {
+      "epoch": 0.19,
+      "eval_accuracy": 0.578970986434046,
+      "eval_loss": 2.444335699081421,
+      "eval_runtime": 87.5675,
+      "eval_samples_per_second": 177.292,
+      "eval_steps_per_second": 22.166,
+      "step": 1500
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 9.140671994500302e-06,
+      "loss": 2.582,
+      "step": 2000
+    },
+    {
+      "epoch": 0.26,
+      "eval_accuracy": 0.5868782143731802,
+      "eval_loss": 2.3700673580169678,
+      "eval_runtime": 83.7436,
+      "eval_samples_per_second": 185.387,
+      "eval_steps_per_second": 23.178,
+      "step": 2000
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 8.926269657128126e-06,
+      "loss": 2.5267,
+      "step": 2500
+    },
+    {
+      "epoch": 0.32,
+      "eval_accuracy": 0.5937291646823507,
+      "eval_loss": 2.309689998626709,
+      "eval_runtime": 81.3517,
+      "eval_samples_per_second": 190.838,
+      "eval_steps_per_second": 23.859,
+      "step": 2500
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 8.711437655753203e-06,
+      "loss": 2.4722,
+      "step": 3000
+    },
+    {
+      "epoch": 0.39,
+      "eval_accuracy": 0.5985969269659629,
+      "eval_loss": 2.2695114612579346,
+      "eval_runtime": 87.4381,
+      "eval_samples_per_second": 177.554,
+      "eval_steps_per_second": 22.199,
+      "step": 3000
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 8.497035318381027e-06,
+      "loss": 2.4289,
+      "step": 3500
+    },
+    {
+      "epoch": 0.45,
+      "eval_accuracy": 0.602404170197503,
+      "eval_loss": 2.2328779697418213,
+      "eval_runtime": 83.7759,
+      "eval_samples_per_second": 185.316,
+      "eval_steps_per_second": 23.169,
+      "step": 3500
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 8.282203317006102e-06,
+      "loss": 2.404,
+      "step": 4000
+    },
+    {
+      "epoch": 0.52,
+      "eval_accuracy": 0.6055254061674608,
+      "eval_loss": 2.206317901611328,
+      "eval_runtime": 87.3965,
+      "eval_samples_per_second": 177.639,
+      "eval_steps_per_second": 22.209,
+      "step": 4000
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 8.067371315631177e-06,
+      "loss": 2.3826,
+      "step": 4500
+    },
+    {
+      "epoch": 0.58,
+      "eval_accuracy": 0.6086694296803393,
+      "eval_loss": 2.183983087539673,
+      "eval_runtime": 87.0314,
+      "eval_samples_per_second": 178.384,
+      "eval_steps_per_second": 22.302,
+      "step": 4500
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 7.852539314256252e-06,
+      "loss": 2.3633,
+      "step": 5000
+    },
+    {
+      "epoch": 0.64,
+      "eval_accuracy": 0.6108753723178051,
+      "eval_loss": 2.1645586490631104,
+      "eval_runtime": 83.9383,
+      "eval_samples_per_second": 184.957,
+      "eval_steps_per_second": 23.124,
+      "step": 5000
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 7.637707312881327e-06,
+      "loss": 2.3425,
+      "step": 5500
+    },
+    {
+      "epoch": 0.71,
+      "eval_accuracy": 0.6121162378522405,
+      "eval_loss": 2.155695676803589,
+      "eval_runtime": 87.4417,
+      "eval_samples_per_second": 177.547,
+      "eval_steps_per_second": 22.198,
+      "step": 5500
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 7.4228753115064025e-06,
+      "loss": 2.333,
+      "step": 6000
+    },
+    {
+      "epoch": 0.77,
+      "eval_accuracy": 0.6140775893820937,
+      "eval_loss": 2.1349785327911377,
+      "eval_runtime": 85.1022,
+      "eval_samples_per_second": 182.428,
+      "eval_steps_per_second": 22.808,
+      "step": 6000
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 7.208472974134228e-06,
+      "loss": 2.311,
+      "step": 6500
+    },
+    {
+      "epoch": 0.84,
+      "eval_accuracy": 0.6151508455851109,
+      "eval_loss": 2.1292011737823486,
+      "eval_runtime": 79.4597,
+      "eval_samples_per_second": 195.382,
+      "eval_steps_per_second": 24.427,
+      "step": 6500
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 6.993640972759303e-06,
+      "loss": 2.3014,
+      "step": 7000
+    },
+    {
+      "epoch": 0.9,
+      "eval_accuracy": 0.6166432908599604,
+      "eval_loss": 2.1181797981262207,
+      "eval_runtime": 87.6275,
+      "eval_samples_per_second": 177.17,
+      "eval_steps_per_second": 22.151,
+      "step": 7000
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 6.7788089713843775e-06,
+      "loss": 2.2974,
+      "step": 7500
+    },
+    {
+      "epoch": 0.97,
+      "eval_accuracy": 0.6169897785349233,
+      "eval_loss": 2.112070083618164,
+      "eval_runtime": 83.9336,
+      "eval_samples_per_second": 184.968,
+      "eval_steps_per_second": 23.125,
+      "step": 7500
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 6.563976970009453e-06,
+      "loss": 2.2866,
+      "step": 8000
+    },
+    {
+      "epoch": 1.03,
+      "eval_accuracy": 0.6173022781800038,
+      "eval_loss": 2.107919454574585,
+      "eval_runtime": 82.2636,
+      "eval_samples_per_second": 188.723,
+      "eval_steps_per_second": 23.595,
+      "step": 8000
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 6.349574632637278e-06,
+      "loss": 2.2675,
+      "step": 8500
+    },
+    {
+      "epoch": 1.1,
+      "eval_accuracy": 0.6191927234863566,
+      "eval_loss": 2.0939817428588867,
+      "eval_runtime": 87.5998,
+      "eval_samples_per_second": 177.226,
+      "eval_steps_per_second": 22.158,
+      "step": 8500
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 6.134742631262354e-06,
+      "loss": 2.2789,
+      "step": 9000
+    },
+    {
+      "epoch": 1.16,
+      "eval_accuracy": 0.6201220093575694,
+      "eval_loss": 2.088168144226074,
+      "eval_runtime": 83.772,
+      "eval_samples_per_second": 185.324,
+      "eval_steps_per_second": 23.17,
+      "step": 9000
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 5.919910629887429e-06,
+      "loss": 2.2684,
+      "step": 9500
+    },
+    {
+      "epoch": 1.22,
+      "eval_accuracy": 0.6199849943877651,
+      "eval_loss": 2.0872652530670166,
+      "eval_runtime": 87.4418,
+      "eval_samples_per_second": 177.547,
+      "eval_steps_per_second": 22.198,
+      "step": 9500
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 5.705078628512504e-06,
+      "loss": 2.2608,
+      "step": 10000
+    },
+    {
+      "epoch": 1.29,
+      "eval_accuracy": 0.6208952330586832,
+      "eval_loss": 2.0795998573303223,
+      "eval_runtime": 86.9343,
+      "eval_samples_per_second": 178.583,
+      "eval_steps_per_second": 22.327,
+      "step": 10000
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 5.490246627137579e-06,
+      "loss": 2.2478,
+      "step": 10500
+    },
+    {
+      "epoch": 1.35,
+      "eval_accuracy": 0.620409766315376,
+      "eval_loss": 2.082674503326416,
+      "eval_runtime": 84.0547,
+      "eval_samples_per_second": 184.701,
+      "eval_steps_per_second": 23.092,
+      "step": 10500
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 5.275844289765404e-06,
+      "loss": 2.2524,
+      "step": 11000
+    },
+    {
+      "epoch": 1.42,
+      "eval_accuracy": 0.6214935816878795,
+      "eval_loss": 2.074056386947632,
+      "eval_runtime": 87.5237,
+      "eval_samples_per_second": 177.381,
+      "eval_steps_per_second": 22.177,
+      "step": 11000
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 5.061012288390479e-06,
+      "loss": 2.2502,
+      "step": 11500
+    },
+    {
+      "epoch": 1.48,
+      "eval_accuracy": 0.6220323169678965,
+      "eval_loss": 2.068490505218506,
+      "eval_runtime": 84.958,
+      "eval_samples_per_second": 182.737,
+      "eval_steps_per_second": 22.847,
+      "step": 11500
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 4.8461802870155545e-06,
+      "loss": 2.243,
+      "step": 12000
+    },
+    {
+      "epoch": 1.55,
+      "eval_accuracy": 0.622761702720804,
+      "eval_loss": 2.0664761066436768,
+      "eval_runtime": 79.0021,
+      "eval_samples_per_second": 196.514,
+      "eval_steps_per_second": 24.569,
+      "step": 12000
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 4.631348285640629e-06,
+      "loss": 2.2417,
+      "step": 12500
+    },
+    {
+      "epoch": 1.61,
+      "eval_accuracy": 0.6228723852166125,
+      "eval_loss": 2.0631983280181885,
+      "eval_runtime": 87.1566,
+      "eval_samples_per_second": 178.128,
+      "eval_steps_per_second": 22.27,
+      "step": 12500
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 4.416516284265704e-06,
+      "loss": 2.2398,
+      "step": 13000
+    },
+    {
+      "epoch": 1.68,
+      "eval_accuracy": 0.6232123058100858,
+      "eval_loss": 2.0592522621154785,
+      "eval_runtime": 83.668,
+      "eval_samples_per_second": 185.555,
+      "eval_steps_per_second": 23.199,
+      "step": 13000
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 4.20168428289078e-06,
+      "loss": 2.2233,
+      "step": 13500
+    },
+    {
+      "epoch": 1.74,
+      "eval_accuracy": 0.6232258668129607,
+      "eval_loss": 2.060002326965332,
+      "eval_runtime": 80.0466,
+      "eval_samples_per_second": 193.95,
+      "eval_steps_per_second": 24.248,
+      "step": 13500
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 3.987281945518604e-06,
+      "loss": 2.2277,
+      "step": 14000
+    },
+    {
+      "epoch": 1.8,
+      "eval_accuracy": 0.623606800420627,
+      "eval_loss": 2.0534963607788086,
+      "eval_runtime": 87.4565,
+      "eval_samples_per_second": 177.517,
+      "eval_steps_per_second": 22.194,
+      "step": 14000
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 3.77244994414368e-06,
+      "loss": 2.2344,
+      "step": 14500
+    },
+    {
+      "epoch": 1.87,
+      "eval_accuracy": 0.6247527084114421,
+      "eval_loss": 2.0484962463378906,
+      "eval_runtime": 83.8183,
+      "eval_samples_per_second": 185.222,
+      "eval_steps_per_second": 23.157,
+      "step": 14500
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 3.5576179427687554e-06,
+      "loss": 2.2274,
+      "step": 15000
+    },
+    {
+      "epoch": 1.93,
+      "eval_accuracy": 0.6244717527399175,
+      "eval_loss": 2.050738573074341,
+      "eval_runtime": 87.5865,
+      "eval_samples_per_second": 177.253,
+      "eval_steps_per_second": 22.161,
+      "step": 15000
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.34321560539658e-06,
+      "loss": 2.2212,
+      "step": 15500
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.6256074101917349,
+      "eval_loss": 2.0428130626678467,
+      "eval_runtime": 86.8032,
+      "eval_samples_per_second": 178.853,
+      "eval_steps_per_second": 22.361,
+      "step": 15500
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 3.1283836040216555e-06,
+      "loss": 2.214,
+      "step": 16000
+    },
+    {
+      "epoch": 2.06,
+      "eval_accuracy": 0.6244417876710062,
+      "eval_loss": 2.0463979244232178,
+      "eval_runtime": 84.1399,
+      "eval_samples_per_second": 184.514,
+      "eval_steps_per_second": 23.069,
+      "step": 16000
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 2.9135516026467303e-06,
+      "loss": 2.2104,
+      "step": 16500
+    },
+    {
+      "epoch": 2.13,
+      "eval_accuracy": 0.6249873550076295,
+      "eval_loss": 2.0476861000061035,
+      "eval_runtime": 87.5417,
+      "eval_samples_per_second": 177.344,
+      "eval_steps_per_second": 22.172,
+      "step": 16500
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 2.698719601271806e-06,
+      "loss": 2.2185,
+      "step": 17000
+    },
+    {
+      "epoch": 2.19,
+      "eval_accuracy": 0.6257313721221357,
+      "eval_loss": 2.039674758911133,
+      "eval_runtime": 84.986,
+      "eval_samples_per_second": 182.677,
+      "eval_steps_per_second": 22.839,
+      "step": 17000
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 2.483887599896881e-06,
+      "loss": 2.2157,
+      "step": 17500
+    },
+    {
+      "epoch": 2.26,
+      "eval_accuracy": 0.6257406865679764,
+      "eval_loss": 2.041879177093506,
+      "eval_runtime": 79.7413,
+      "eval_samples_per_second": 194.692,
+      "eval_steps_per_second": 24.341,
+      "step": 17500
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 2.2690555985219558e-06,
+      "loss": 2.2128,
+      "step": 18000
+    },
+    {
+      "epoch": 2.32,
+      "eval_accuracy": 0.6254893845927666,
+      "eval_loss": 2.043928623199463,
+      "eval_runtime": 87.45,
+      "eval_samples_per_second": 177.53,
+      "eval_steps_per_second": 22.196,
+      "step": 18000
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 2.054223597147031e-06,
+      "loss": 2.2154,
+      "step": 18500
+    },
+    {
+      "epoch": 2.38,
+      "eval_accuracy": 0.6259225237275015,
+      "eval_loss": 2.037231683731079,
+      "eval_runtime": 83.6819,
+      "eval_samples_per_second": 185.524,
+      "eval_steps_per_second": 23.195,
+      "step": 18500
+    },
+    {
+      "epoch": 2.45,
+      "learning_rate": 1.8393915957721066e-06,
+      "loss": 2.2099,
+      "step": 19000
+    },
+    {
+      "epoch": 2.45,
+      "eval_accuracy": 0.62631184758297,
+      "eval_loss": 2.0336806774139404,
+      "eval_runtime": 81.3506,
+      "eval_samples_per_second": 190.841,
+      "eval_steps_per_second": 23.86,
+      "step": 19000
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 1.6245595943971814e-06,
+      "loss": 2.2045,
+      "step": 19500
+    },
+    {
+      "epoch": 2.51,
+      "eval_accuracy": 0.6258799592390727,
+      "eval_loss": 2.039562225341797,
+      "eval_runtime": 87.4501,
+      "eval_samples_per_second": 177.53,
+      "eval_steps_per_second": 22.196,
+      "step": 19500
+    },
+    {
+      "epoch": 2.58,
+      "learning_rate": 1.4097275930222567e-06,
+      "loss": 2.2138,
+      "step": 20000
+    },
+    {
+      "epoch": 2.58,
+      "eval_accuracy": 0.6261649440028011,
+      "eval_loss": 2.0390186309814453,
+      "eval_runtime": 83.8434,
+      "eval_samples_per_second": 185.167,
+      "eval_steps_per_second": 23.15,
+      "step": 20000
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 1.194895591647332e-06,
+      "loss": 2.2103,
+      "step": 20500
+    },
+    {
+      "epoch": 2.64,
+      "eval_accuracy": 0.6262993215315168,
+      "eval_loss": 2.03385329246521,
+      "eval_runtime": 87.3376,
+      "eval_samples_per_second": 177.759,
+      "eval_steps_per_second": 22.224,
+      "step": 20500
+    },
+    {
+      "epoch": 2.64,
+      "step": 20500,
+      "total_flos": 1.0082485751267328e+16,
+      "train_loss": 2.351401915015244,
+      "train_runtime": 7059.8447,
+      "train_samples_per_second": 52.745,
+      "train_steps_per_second": 3.297
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 23274,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 500,
+  "total_flos": 1.0082485751267328e+16,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}