Model save

README.md

@@ -1,11 +1,8 @@
 ---
 license: apache-2.0
-base_model: bert-base-uncased
+base_model: distilbert-base-uncased
 tags:
-- question-answering
 - generated_from_trainer
-datasets:
-- squad
 model-index:
 - name: out
   results: []
@@ -16,7 +13,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # out
 
-This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the squad dataset.
+This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the None dataset.
 
 ## Model description
 
@@ -36,12 +33,12 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 5e-05
-- train_batch_size: 32
+- train_batch_size: 256
 - eval_batch_size: 16
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 20.0
+- num_epochs: 3.0
 
 ### Training results
 

config.json

@@ -1,26 +1,24 @@
 {
-  "_name_or_path": "bert-base-uncased",
+  "_name_or_path": "distilbert-base-uncased",
+  "activation": "gelu",
   "architectures": [
-    "BertForQuestionAnswering"
+    "DistilBertForQuestionAnswering"
   ],
-  "attention_probs_dropout_prob": 0.1,
-  "classifier_dropout": null,
-  "gradient_checkpointing": false,
-  "hidden_act": "gelu",
-  "hidden_dropout_prob": 0.1,
-  "hidden_size": 768,
+  "attention_dropout": 0.1,
+  "dim": 768,
+  "dropout": 0.1,
+  "hidden_dim": 3072,
   "initializer_range": 0.02,
-  "intermediate_size": 3072,
-  "layer_norm_eps": 1e-12,
   "max_position_embeddings": 512,
-  "model_type": "bert",
-  "num_attention_heads": 12,
-  "num_hidden_layers": 12,
+  "model_type": "distilbert",
+  "n_heads": 12,
+  "n_layers": 6,
   "pad_token_id": 0,
-  "position_embedding_type": "absolute",
+  "qa_dropout": 0.1,
+  "seq_classif_dropout": 0.2,
+  "sinusoidal_pos_embds": false,
+  "tie_weights_": true,
   "torch_dtype": "float32",
   "transformers_version": "4.35.0.dev0",
-  "type_vocab_size": 2,
-  "use_cache": true,
   "vocab_size": 30522
 }

pruning_config.json

@@ -18,7 +18,7 @@
   },
   "pruning_fn": "unstructured_magnitude",
   "pruning_fn_default_kwargs": {
-    "target_sparsity": 0.85
+    "target_sparsity": 0.9
   },
   "pruning_frequency": 100,
   "scheduler": "iterative",

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bac485f19250c36d787180971157213cfd2ea1b3afce31dc71f153a39bf69a9a
-size 775405638
+oid sha256:45f49d0d4ef9d2b6a142d56f5151687051e5e54606390c6de709e339848665c4
+size 265493026

raw_pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:41ae827d2be311931b38e6f9b4c2bebf1c47fd051102986fbdd25a39daf39e7b
-size 775420626
+oid sha256:ace5d339b09a1b5d89214c499f75ccb2fefdcce154e576b06cc71a1361c8eb03
+size 435382250

tokenizer_config.json

@@ -51,6 +51,6 @@
   "sep_token": "[SEP]",
   "strip_accents": null,
   "tokenize_chinese_chars": true,
-  "tokenizer_class": "BertTokenizer",
+  "tokenizer_class": "DistilBertTokenizer",
   "unk_token": "[UNK]"
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4c5d2f0a58150c0b62d9c0ed63d20b066a023879d3bd3b541d01997c584f2d77
+oid sha256:e9b02eee87a0bc35b4c8b5e908bd256e0003991234aed44f5498f04da284bb98
 size 4664