Upload MiniMaxM2ForCausalLM

config.json

@@ -4,20 +4,20 @@
   ],
   "attention_dropout": 0.0,
   "bos_token_id": 200034,
-  "dtype": "bfloat16",
+  "dtype": "float32",
   "eos_token_id": 200020,
   "head_dim": 128,
   "hidden_act": "silu",
-  "hidden_size": 3072,
+  "hidden_size": 256,
   "initializer_range": 0.02,
-  "intermediate_size": 1536,
+  "intermediate_size": 128,
   "max_position_embeddings": 196608,
   "model_type": "minimax_m2",
-  "num_attention_heads": 48,
-  "num_experts_per_tok": 8,
+  "num_attention_heads": 2,
+  "num_experts_per_tok": 2,
   "num_hidden_layers": 2,
-  "num_key_value_heads": 8,
-  "num_local_experts": 256,
+  "num_key_value_heads": 1,
+  "num_local_experts": 8,
   "output_router_logits": false,
   "rms_norm_eps": 1e-06,
   "rope_parameters": {
@@ -25,9 +25,10 @@
     "rope_theta": 5000000.0,
     "rope_type": "default"
   },
+  "rotary_dim": 64,
   "router_aux_loss_coef": 0.001,
   "router_jitter_noise": 0.0,
-  "tie_word_embeddings": false,
+  "tie_word_embeddings": true,
   "transformers_version": "5.0.0.dev0",
   "use_cache": true,
   "vocab_size": 200064

generation_config.json

@@ -1,8 +1,6 @@
 {
-  "bos_token_id": 200019,
-  "do_sample": true,
+  "_from_model_config": true,
+  "bos_token_id": 200034,
   "eos_token_id": 200020,
-  "top_k": 40,
-  "top_p": 0.95,
   "transformers_version": "5.0.0.dev0"
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:080c65adaeede0c0c74988f90c6df73be5bdb154f4eee08c99df0b10e5c122c7
-size 17133474240
+oid sha256:d4a2a1ce27a61455d3d40e607ab450ed7a3d8aafd5cc436fe6f84ddb023578c9
+size 212762968