Model save

README.md

@@ -15,8 +15,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model was trained from scratch on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.9785
-- Accuracy: 0.4472
+- Loss: 4.2249
+- Accuracy: 0.0948
 
 ## Model description
 
@@ -36,8 +36,8 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 5e-05
-- train_batch_size: 8
-- eval_batch_size: 8
+- train_batch_size: 4
+- eval_batch_size: 4
 - seed: 123
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
@@ -48,11 +48,11 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step   | Validation Loss | Accuracy |
 |:-------------:|:-----:|:------:|:---------------:|:--------:|
-| 2.0838        | 1.0   | 34200  | 2.1737          | 0.4064   |
-| 1.8703        | 2.0   | 68400  | 2.0177          | 0.4362   |
-| 1.8733        | 3.0   | 102600 | 1.9902          | 0.4368   |
-| 1.6634        | 4.0   | 136800 | 1.9717          | 0.4455   |
-| 1.6316        | 5.0   | 171000 | 1.9785          | 0.4472   |
+| 4.1776        | 1.0   | 68400  | 4.2204          | 0.0948   |
+| 4.2058        | 2.0   | 136800 | 4.2118          | 0.0948   |
+| 4.1949        | 3.0   | 205200 | 4.2219          | 0.0948   |
+| 4.1297        | 4.0   | 273600 | 4.2298          | 0.0948   |
+| 4.2056        | 5.0   | 342000 | 4.2249          | 0.0948   |
 
 
 ### Framework versions

config.json

@@ -8,7 +8,7 @@
   "gradient_checkpointing": false,
   "hidden_act": "gelu",
   "hidden_dropout_prob": 0.1,
-  "hidden_size": 768,
+  "hidden_size": 1024,
   "id2label": {
     "0": "LABEL_0",
     "1": "LABEL_1",
@@ -212,7 +212,7 @@
     "199": "LABEL_199"
   },
   "initializer_range": 0.02,
-  "intermediate_size": 3072,
+  "intermediate_size": 4096,
   "label2id": {
     "LABEL_0": 0,
     "LABEL_1": 1,
@@ -418,8 +418,8 @@
   "layer_norm_eps": 1e-12,
   "max_position_embeddings": 512,
   "model_type": "bert",
-  "num_attention_heads": 12,
-  "num_hidden_layers": 12,
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
   "pad_token_id": 0,
   "position_embedding_type": "absolute",
   "problem_type": "single_label_classification",

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7c5ae12f57547f1a771d7d67e797e2af145391866394d8dfa8168a08f8fe1691
-size 473551648
+oid sha256:9c7d0048dc65573b7291b2036889f6e18b0345c3c3111fff8eec59693017cdec
+size 1388079840

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fd41aa8a858ba1c9475c1b00f49df12f3c431f5e41d29eed4d95382544c7d510
+oid sha256:51e423a43e2c343212f9e7679f0b3d1aff9fe7a7b292399f81c25b5157e5e97d
 size 4664