初始化项目，由ModelHub XC社区提供模型

Model: vaiv/GeM2-Llamion-14B-LongChat Source: Original Platform
2026-07-16 19:37:52 +08:00
commit 31f500fd7f
17 changed files with 170447 additions and 0 deletions
--- a/config.json
+++ b/config.json
@@ -0,0 +1,31 @@
+{
+  "_name_or_path": "vaiv/GeM2-Llamion-14B-LongChat",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 5120,
+  "initializer_range": 0.02,
+  "intermediate_size": 15360,
+  "max_position_embeddings": 200000,
+  "max_sequence_length": 200000,
+  "mlp_bias": false,
+  "model_type": "llama",
+  "num_attention_heads": 40,
+  "num_hidden_layers": 40,
+  "num_key_value_heads": 40,
+  "pad_token_id": 0,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 50000000.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.41.0",
+  "use_cache": false,
+  "vocab_size": 84608
+}