Upload DogeForCausalLM

config.json

@@ -1,47 +1,47 @@
-{
-  "_name_or_path": "./results/Doge-60M-Instruct-SFT",
-  "architectures": [
-    "DogeForCausalLM"
-  ],
-  "attention_dropout": 0.0,
-  "auto_map": {
-    "AutoConfig": "configuration_doge.DogeConfig",
-    "AutoModelForCausalLM": "modeling_doge.DogeForCausalLM"
-  },
-  "bos_token_id": 0,
-  "dynamic_mask_ratio": 0.0,
-  "eos_token_id": 1,
-  "expert_retrieval_size": 256,
-  "hidden_act": "silu",
-  "hidden_bias": false,
-  "hidden_dropout": 0.0,
-  "hidden_size": 512,
-  "initializer_range": 0.02,
-  "intermediate_size": 1024,
-  "is_moe": false,
-  "max_position_embeddings": 2048,
-  "model_type": "doge",
-  "num_attention_heads": 4,
-  "num_cdmmoe_experts": 2048,
-  "num_cdmmoe_experts_per_head": 8,
-  "num_cdmmoe_heads": 4,
-  "num_cdmoe_experts": 16348,
-  "num_cdmoe_experts_per_head": 8,
-  "num_cdmoe_heads": 4,
-  "num_channels": 3,
-  "num_hidden_layers": 16,
-  "num_key_value_heads": 2,
-  "pad_token_id": 2,
-  "patch_size": 16,
-  "rms_norm_eps": 1e-06,
-  "rope_scaling": {
-    "factor": 4.0,
-    "original_max_position_embeddings": 2048,
-    "rope_type": "dynamic"
-  },
-  "rope_theta": 10000.0,
-  "torch_dtype": "float32",
-  "transformers_version": "4.49.0.dev0",
-  "use_cache": true,
-  "vocab_size": 32768
-}
+{
+  "_name_or_path": "./results/Doge-20M-Instruct-SFT",
+  "architectures": [
+    "DogeForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "auto_map": {
+    "AutoConfig": "configuration_doge.DogeConfig",
+    "AutoModelForCausalLM": "modeling_doge.DogeForCausalLM"
+  },
+  "bos_token_id": 0,
+  "dynamic_mask_ratio": 0.0,
+  "eos_token_id": 1,
+  "expert_retrieval_size": 256,
+  "hidden_act": "silu",
+  "hidden_bias": false,
+  "hidden_dropout": 0.0,
+  "hidden_size": 256,
+  "initializer_range": 0.02,
+  "intermediate_size": 512,
+  "is_moe": false,
+  "max_position_embeddings": 2048,
+  "model_type": "doge",
+  "num_attention_heads": 2,
+  "num_cdmmoe_experts": 2048,
+  "num_cdmmoe_experts_per_head": 8,
+  "num_cdmmoe_heads": 4,
+  "num_cdmoe_experts": 16348,
+  "num_cdmoe_experts_per_head": 8,
+  "num_cdmoe_heads": 4,
+  "num_channels": 3,
+  "num_hidden_layers": 8,
+  "num_key_value_heads": 1,
+  "pad_token_id": 2,
+  "patch_size": 16,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": {
+    "factor": 4.0,
+    "original_max_position_embeddings": 2048,
+    "rope_type": "dynamic"
+  },
+  "rope_theta": 10000.0,
+  "torch_dtype": "float32",
+  "transformers_version": "4.48.1",
+  "use_cache": true,
+  "vocab_size": 32768
+}

generation_config.json

@@ -1,7 +1,7 @@
-{
-  "_from_model_config": true,
-  "bos_token_id": 0,
-  "eos_token_id": 1,
-  "pad_token_id": 2,
-  "transformers_version": "4.49.0.dev0"
-}
+{
+  "_from_model_config": true,
+  "bos_token_id": 0,
+  "eos_token_id": 1,
+  "pad_token_id": 2,
+  "transformers_version": "4.48.1"
+}

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d00675b5448609e1d7f939db78b11b6556d4c4ae5b2908ad2fb5a5a1989baa72
-size 218325576
+oid sha256:15a80619903e15f83a9f9982e171d5508325e17317ac905186e89f839f83b330
+size 52482152