Upload MiniMaxM2ForCausalLM
Browse files- config.json +1 -2
- model.safetensors +2 -2
config.json
CHANGED
|
@@ -15,7 +15,7 @@
|
|
| 15 |
"model_type": "minimax_m2",
|
| 16 |
"num_attention_heads": 2,
|
| 17 |
"num_experts_per_tok": 2,
|
| 18 |
-
"num_hidden_layers":
|
| 19 |
"num_key_value_heads": 1,
|
| 20 |
"num_local_experts": 8,
|
| 21 |
"output_router_logits": false,
|
|
@@ -25,7 +25,6 @@
|
|
| 25 |
"rope_theta": 5000000.0,
|
| 26 |
"rope_type": "default"
|
| 27 |
},
|
| 28 |
-
"rotary_dim": 64,
|
| 29 |
"router_aux_loss_coef": 0.001,
|
| 30 |
"router_jitter_noise": 0.0,
|
| 31 |
"tie_word_embeddings": true,
|
|
|
|
| 15 |
"model_type": "minimax_m2",
|
| 16 |
"num_attention_heads": 2,
|
| 17 |
"num_experts_per_tok": 2,
|
| 18 |
+
"num_hidden_layers": 12,
|
| 19 |
"num_key_value_heads": 1,
|
| 20 |
"num_local_experts": 8,
|
| 21 |
"output_router_logits": false,
|
|
|
|
| 25 |
"rope_theta": 5000000.0,
|
| 26 |
"rope_type": "default"
|
| 27 |
},
|
|
|
|
| 28 |
"router_aux_loss_coef": 0.001,
|
| 29 |
"router_jitter_noise": 0.0,
|
| 30 |
"tie_word_embeddings": true,
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1f85041ccc423ed70db486dd07d66537d60beeccde5b0c5d0c14e0ff25f303ee
|
| 3 |
+
size 252243968
|