Add config.json

config.json

@@ -0,0 +1,58 @@
+{
+  "model_type": "transformer",
+  "architectures": ["Sheikh2_5CoderForCausalLM"],
+  "max_position_embeddings": 32768,
+  "vocab_size": 50257,
+  "hidden_size": 3072,
+  "num_attention_heads": 16,
+  "num_key_value_heads": 2,
+  "num_hidden_layers": 36,
+  "intermediate_size": 8192,
+  "pad_token_id": 0,
+  "eos_token_id": 1,
+  "bos_token_id": 2,
+  "rope_theta": 10000.0,
+  "rope_scaling": {
+    "type": "linear",
+    "factor": 8.0
+  },
+  "use_cache": true,
+  "tie_word_embeddings": true,
+  "layer_norm_epsilon": 1e-6,
+  "mlp_bias": false,
+  "attention_bias": true,
+  "qkv_proj_bias": true,
+  "rms_norm_eps": 1e-6,
+  "activation_function": "swiglu",
+  "torch_dtype": "bfloat16",
+  "pretraining_tp": 1,
+  "reduction_factor": 32,
+  "num_experts_per_tok": 2,
+  "num_local_experts": 8,
+  "model_name": "Sheikh-2.5-Coder",
+  "model_version": "1.0.0",
+  "training_objectives": [
+    "causal_language_modeling",
+    "instruction_tuning",
+    "code_generation"
+  ],
+  "supported_languages": [
+    "python",
+    "javascript",
+    "typescript",
+    "java",
+    "cpp",
+    "c",
+    "go",
+    "rust",
+    "php",
+    "ruby",
+    "swift",
+    "kotlin",
+    "scala",
+    "r",
+    "sql",
+    "html",
+    "css"
+  ]
+}