Training in progress, step 500

config.json

@@ -1,17 +1,13 @@
 {
   "_name_or_path": "AIRI-Institute/gena-lm-bert-large-t2t",
   "architectures": [
-    "BigBirdForSequenceClassification"
+    "BertForSequenceClassification"
   ],
   "attention_probs_dropout_prob": 0.1,
-  "attention_type": "block_sparse",
   "auto_map": {
     "AutoModel": "AIRI-Institute/gena-lm-bert-large-t2t--modeling_bert.BertForMaskedLM"
   },
-  "block_size": 64,
-  "bos_token_id": 1,
   "classifier_dropout": null,
-  "eos_token_id": 2,
   "gradient_checkpointing": false,
   "hidden_act": "gelu",
   "hidden_dropout_prob": 0.1,
@@ -20,20 +16,16 @@
   "intermediate_size": 4096,
   "layer_norm_eps": 1e-05,
   "max_position_embeddings": 512,
-  "model_type": "big_bird",
+  "model_type": "bert",
   "num_attention_heads": 16,
   "num_hidden_layers": 24,
-  "num_random_blocks": 3,
   "pad_token_id": 3,
   "position_embedding_type": "absolute",
   "pre_layer_norm": true,
   "problem_type": "single_label_classification",
-  "rescale_embeddings": false,
-  "sep_token_id": 66,
   "torch_dtype": "float32",
   "transformers_version": "4.42.0.dev0",
   "type_vocab_size": 2,
-  "use_bias": true,
   "use_cache": true,
   "vocab_size": 32000
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0fdfc85838592e72851401bea7e235199a631da31f6a4ef141e1be13fef66ae3
-size 1350875256
+oid sha256:7609ebfac4eccb80579fe6e8e7ef0c61d80c79502f34403a7e929c2480337226
+size 1346688816