change config.json

config.json

@@ -1,12 +1,11 @@
 {
-  "_name_or_path": "/home/kas/data/llm/qwen/",
   "architectures": [
-    "QWenLMHeadModel"
+    "MonkeyLMHeadModel"
   ],
   "attn_dropout_prob": 0.0,
   "auto_map": {
     "AutoConfig": "configuration_qwen.QWenConfig",
-    "AutoModelForCausalLM": "modeling_qwen.QWenLMHeadModel"
+    "AutoModelForCausalLM": "modeling_monkey.MonkeyLMHeadModel"
   },
   "bf16": true,
   "emb_dropout_prob": 0.0,
@@ -18,7 +17,7 @@
   "kv_channels": 128,
   "layer_norm_epsilon": 1e-06,
   "max_position_embeddings": 8192,
-  "model_type": "qwen",
+  "model_type": "monkey",
   "no_bias": true,
   "num_attention_heads": 32,
   "num_hidden_layers": 32,
@@ -36,19 +35,15 @@
   "use_flash_attn": false,
   "use_logn_attn": true,
   "visual": {
-    "add_window": false,
     "heads": 16,
     "image_size": 896,
     "image_start_id": 151857,
     "layers": 48,
-    "lora_repeat_num": 4,
     "mlp_ratio": 4.9231,
-    "n_queries": 256,
     "output_dim": 4096,
     "patch_size": 14,
-    "use_global": true,
     "width": 1664,
-    "window_all": false
+    "lora_repeat_num":4
   },
   "vocab_size": 151936
 }

modeling_monkey.py

@@ -26,7 +26,7 @@ try:
 except ImportError:
     rearrange = None
 from torch import nn
-from monkey_model.modeling_qwen import QWenModel,QWenPreTrainedModel,QWenLMHeadModel
+from .modeling_qwen import QWenModel,QWenPreTrainedModel,QWenLMHeadModel
 SUPPORT_CUDA = torch.cuda.is_available()
 SUPPORT_BF16 = SUPPORT_CUDA and torch.cuda.is_bf16_supported()
 SUPPORT_FP16 = SUPPORT_CUDA and torch.cuda.get_device_capability(0)[0] >= 7

training_args.bin

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:60d8ad6ecdf0d6dc385ea49113e3c6c6ace92ef3a6f3c9707e5b528a592734d2
-size 5563