Upload Qwen3MoeForSequenceClassification
Browse files- config.json +3 -3
- model.safetensors +2 -2
config.json
CHANGED
|
@@ -15,7 +15,7 @@
|
|
| 15 |
"0": "LABEL_0"
|
| 16 |
},
|
| 17 |
"initializer_range": 0.02,
|
| 18 |
-
"intermediate_size":
|
| 19 |
"label2id": {
|
| 20 |
"LABEL_0": 0
|
| 21 |
},
|
|
@@ -29,7 +29,7 @@
|
|
| 29 |
"num_experts": 4,
|
| 30 |
"num_experts_per_tok": 2,
|
| 31 |
"num_hidden_layers": 2,
|
| 32 |
-
"num_key_value_heads":
|
| 33 |
"output_router_logits": false,
|
| 34 |
"rms_norm_eps": 1e-06,
|
| 35 |
"rope_scaling": null,
|
|
@@ -37,7 +37,7 @@
|
|
| 37 |
"router_aux_loss_coef": 0.001,
|
| 38 |
"sliding_window": null,
|
| 39 |
"tie_word_embeddings": false,
|
| 40 |
-
"transformers_version": "4.
|
| 41 |
"use_cache": true,
|
| 42 |
"use_sliding_window": false,
|
| 43 |
"vocab_size": 151936
|
|
|
|
| 15 |
"0": "LABEL_0"
|
| 16 |
},
|
| 17 |
"initializer_range": 0.02,
|
| 18 |
+
"intermediate_size": 32,
|
| 19 |
"label2id": {
|
| 20 |
"LABEL_0": 0
|
| 21 |
},
|
|
|
|
| 29 |
"num_experts": 4,
|
| 30 |
"num_experts_per_tok": 2,
|
| 31 |
"num_hidden_layers": 2,
|
| 32 |
+
"num_key_value_heads": 2,
|
| 33 |
"output_router_logits": false,
|
| 34 |
"rms_norm_eps": 1e-06,
|
| 35 |
"rope_scaling": null,
|
|
|
|
| 37 |
"router_aux_loss_coef": 0.001,
|
| 38 |
"sliding_window": null,
|
| 39 |
"tie_word_embeddings": false,
|
| 40 |
+
"transformers_version": "4.56.2",
|
| 41 |
"use_cache": true,
|
| 42 |
"use_sliding_window": false,
|
| 43 |
"vocab_size": 151936
|
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:a640b1157874512f1285a05aa8fc4c65e6fe9f06b14ec911cc915428233febba
|
| 3 |
+
size 5556504
|