update model

README.md

@@ -25,7 +25,7 @@ All the models have a comparable model size between 90 MB and 150 MB, BPE tokeni
 ### Model Sources
 
 - **Repository:** [Plant DNA LLMs](https://github.com/zhangtaolab/plant_DNA_LLMs)
-- **Manuscript:** [Versatile applications of foundation DNA large language models in plant genomes]() 
+- **Manuscript:** [PDLLMs: A group of tailored DNA large language models for analyzing plant genomes]() 
 
 ### Architecture
 

config.json

@@ -1 +1,42 @@
-{"d_model": 768, "n_layer": 24, "vocab_size": 8000, "ssm_cfg": {}, "rms_norm": true, "residual_in_fp32": true, "fused_add_norm": true, "pad_vocab_size_multiple": 8, "tie_embeddings": true}
+{
+  "_name_or_path": "Plant_DNAMamba_BPE_conservation",
+  "architectures": [
+    "MambaForSequenceClassification"
+  ],
+  "bos_token_id": 0,
+  "conv_kernel": 4,
+  "d_inner": 1536,
+  "d_model": 768,
+  "eos_token_id": 0,
+  "expand": 2,
+  "fused_add_norm": true,
+  "hidden_act": "silu",
+  "hidden_size": 768,
+  "initializer_range": 0.1,
+  "intermediate_size": 1536,
+  "layer_norm_epsilon": 1e-05,
+  "model_type": "mamba",
+  "n_layer": 24,
+  "num_hidden_layers": 24,
+  "pad_token_id": 0,
+  "pad_vocab_size_multiple": 8,
+  "problem_type": "single_label_classification",
+  "rescale_prenorm_residual": false,
+  "residual_in_fp32": true,
+  "rms_norm": true,
+  "ssm_cfg": {},
+  "state_size": 16,
+  "time_step_floor": 0.0001,
+  "time_step_init_scheme": "random",
+  "time_step_max": 0.1,
+  "time_step_min": 0.001,
+  "time_step_rank": 48,
+  "time_step_scale": 1.0,
+  "torch_dtype": "float32",
+  "transformers_version": "4.45.0.dev0",
+  "use_bias": false,
+  "use_cache": false,
+  "use_conv_bias": true,
+  "use_mambapy": false,
+  "vocab_size": 8000
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2c6ecf6f4c111b50f9fd72a206b6b4d6b4ffa79c7403f9dee9057a26f7b3bf25
+size 386689624

tokenizer.json

@@ -46,10 +46,30 @@
     "use_regex": true
   },
   "post_processor": {
-    "type": "ByteLevel",
-    "add_prefix_space": false,
-    "trim_offsets": true,
-    "use_regex": true
+    "type": "TemplateProcessing",
+    "single": [
+      {
+        "Sequence": {
+          "id": "A",
+          "type_id": 0
+        }
+      }
+    ],
+    "pair": [
+      {
+        "Sequence": {
+          "id": "A",
+          "type_id": 0
+        }
+      },
+      {
+        "Sequence": {
+          "id": "B",
+          "type_id": 1
+        }
+      }
+    ],
+    "special_tokens": {}
   },
   "decoder": {
     "type": "ByteLevel",
@@ -65,6 +85,7 @@
     "end_of_word_suffix": null,
     "fuse_unk": false,
     "byte_fallback": false,
+    "ignore_merges": false,
     "vocab": {
       "<|endoftext|>": 0,
       "<|padding|>": 1,

tokenizer_config.json

@@ -1,4 +1,6 @@
 {
+  "add_bos_token": false,
+  "add_eos_token": false,
   "add_prefix_space": false,
   "added_tokens_decoder": {
     "0": {