Upload folder using huggingface_hub

LLaDA-gpt2-355M-Base-router/added_tokens.json

@@ -0,0 +1,3 @@
+{
+  "<|mdm_mask|>": 50257
+}

LLaDA-gpt2-355M-Base-router/config.json

@@ -0,0 +1,50 @@
+{
+    "activation_type": "silu",
+    "alibi": false,
+    "alibi_bias_max": 8.0,
+    "architectures": [
+        "LLaDAModelLM"
+    ],
+    "attention_dropout": 0.0,
+    "attention_layer_norm": false,
+    "attention_layer_norm_with_affine": true,
+    "bias_for_layer_norm": false,
+    "block_group_size": 1,
+    "block_type": "llama",
+    "d_model": 1024,
+    "embedding_dropout": 0.0,
+    "embedding_size": 50304,
+    "eos_token_id": 50256,
+    "flash_attention": false,
+    "include_bias": false,
+    "include_qkv_bias": false,
+    "init_cutoff_factor": null,
+    "init_device": "meta",
+    "init_fn": "mitchell",
+    "init_std": 0.02,
+    "input_emb_norm": false,
+    "layer_norm_type": "rms",
+    "layer_norm_with_affine": true,
+    "mask_token_id": 50257,
+    "max_sequence_length": 4096,
+    "mlp_hidden_size": 4096,
+    "mlp_ratio": 4,
+    "model_type": "lladarouter",
+    "multi_query_attention": null,
+    "n_heads": 16,
+    "n_kv_heads": 16,
+    "n_layers": 15,
+    "pad_token_id": 50256,
+    "precision": "amp_bf16",
+    "residual_dropout": 0.0,
+    "rms_norm_eps": 1e-05,
+    "rope": true,
+    "rope_full_precision": true,
+    "rope_theta": 500000.0,
+    "scale_logits": false,
+    "transformers_version": "4.46.3",
+    "use_cache": false,
+    "vocab_size": 50304,
+    "weight_tying": false,
+    "n_experts": 4
+}

LLaDA-gpt2-355M-Base-router/generation_config.json

@@ -0,0 +1,6 @@
+{
+    "_from_model_config": true,
+    "bos_token_id": 126080,
+    "eos_token_id": 126081,
+    "transformers_version": "4.38.2"
+}

LLaDA-gpt2-355M-Base-router/special_tokens_map.json

@@ -0,0 +1,13 @@
+{
+  "bos_token": "<|endoftext|>",
+  "eos_token": "<|endoftext|>",
+  "mask_token": {
+    "content": "<|mdm_mask|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|endoftext|>",
+  "unk_token": "<|endoftext|>"
+}

LLaDA-gpt2-355M-Base-router/tokenizer_config.json

@@ -0,0 +1,30 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "50256": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "50257": {
+      "content": "<|mdm_mask|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|endoftext|>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "extra_special_tokens": {},
+  "mask_token": "<|mdm_mask|>",
+  "model_max_length": 1024,
+  "pad_token": "<|endoftext|>",
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": "<|endoftext|>"
+}