Update config.json
Browse files- config.json +13 -60
config.json
CHANGED
|
@@ -1,65 +1,18 @@
|
|
| 1 |
{
|
|
|
|
| 2 |
"architectures": [
|
| 3 |
-
"
|
| 4 |
],
|
| 5 |
-
"
|
| 6 |
-
"
|
| 7 |
-
"
|
| 8 |
-
"
|
| 9 |
-
"
|
| 10 |
-
"
|
| 11 |
-
"
|
| 12 |
-
"
|
| 13 |
-
"
|
| 14 |
-
"layer_types": [
|
| 15 |
-
"full_attention",
|
| 16 |
-
"full_attention",
|
| 17 |
-
"full_attention",
|
| 18 |
-
"full_attention",
|
| 19 |
-
"full_attention",
|
| 20 |
-
"full_attention",
|
| 21 |
-
"full_attention",
|
| 22 |
-
"full_attention",
|
| 23 |
-
"full_attention",
|
| 24 |
-
"full_attention",
|
| 25 |
-
"full_attention",
|
| 26 |
-
"full_attention",
|
| 27 |
-
"full_attention",
|
| 28 |
-
"full_attention",
|
| 29 |
-
"full_attention",
|
| 30 |
-
"full_attention",
|
| 31 |
-
"full_attention",
|
| 32 |
-
"full_attention",
|
| 33 |
-
"full_attention",
|
| 34 |
-
"full_attention",
|
| 35 |
-
"full_attention",
|
| 36 |
-
"full_attention",
|
| 37 |
-
"full_attention",
|
| 38 |
-
"full_attention",
|
| 39 |
-
"full_attention",
|
| 40 |
-
"full_attention",
|
| 41 |
-
"full_attention",
|
| 42 |
-
"full_attention"
|
| 43 |
-
],
|
| 44 |
-
"max_position_embeddings": 40960,
|
| 45 |
-
"max_window_layers": 28,
|
| 46 |
-
"model_type": "qwen3",
|
| 47 |
-
"num_attention_heads": 16,
|
| 48 |
-
"num_hidden_layers": 28,
|
| 49 |
-
"num_key_value_heads": 8,
|
| 50 |
-
"output_hidden_states": true,
|
| 51 |
-
"rms_norm_eps": 1e-06,
|
| 52 |
-
"rope_scaling": null,
|
| 53 |
-
"rope_theta": 1000000,
|
| 54 |
-
"sliding_window": null,
|
| 55 |
-
"tie_word_embeddings": true,
|
| 56 |
"torch_dtype": "bfloat16",
|
| 57 |
-
"
|
| 58 |
-
"
|
| 59 |
-
"use_sliding_window": false,
|
| 60 |
-
"vocab_size": 151936,
|
| 61 |
-
"auto_map": {
|
| 62 |
-
"AutoConfig": "splade.SpladeConfig",
|
| 63 |
-
"AutoModelForCausalLM": "splade.Splade"
|
| 64 |
-
}
|
| 65 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"archi_type": "decoder",
|
| 3 |
"architectures": [
|
| 4 |
+
"Splade"
|
| 5 |
],
|
| 6 |
+
"attn_implementation": "flash_attention_2",
|
| 7 |
+
"attn_type": "causal",
|
| 8 |
+
"bidirectional": true,
|
| 9 |
+
"lora": false,
|
| 10 |
+
"lora_r": 0,
|
| 11 |
+
"model_name_or_path": "Qwen/Qwen3-0.6B",
|
| 12 |
+
"model_type": "splade",
|
| 13 |
+
"n_layers": null,
|
| 14 |
+
"padding_side": "left",
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 15 |
"torch_dtype": "bfloat16",
|
| 16 |
+
"train_head": false,
|
| 17 |
+
"transformers_version": "4.53.3"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 18 |
}
|