Upload folder using huggingface_hub
Browse files- config.json +13 -14
- generation_config.json +6 -4
config.json
CHANGED
|
@@ -8,39 +8,38 @@
|
|
| 8 |
"eos_token_id": 151643,
|
| 9 |
"head_dim": 128,
|
| 10 |
"hidden_act": "silu",
|
| 11 |
-
"hidden_size":
|
| 12 |
"initializer_range": 0.02,
|
| 13 |
-
"intermediate_size":
|
| 14 |
"max_position_embeddings": 131072,
|
| 15 |
-
"max_window_layers":
|
| 16 |
-
"mix_coeff":
|
| 17 |
"model_type": "qwen3",
|
| 18 |
-
"num_attention_heads":
|
| 19 |
-
"num_hidden_layers":
|
| 20 |
-
"num_key_value_heads":
|
| 21 |
"rms_norm_eps": 1e-06,
|
| 22 |
"rope_scaling": null,
|
| 23 |
-
"rope_theta":
|
| 24 |
"sliding_window": null,
|
| 25 |
"tie_word_embeddings": true,
|
| 26 |
"torch_dtype": "bfloat16",
|
| 27 |
"transformers_version": "4.51.0",
|
| 28 |
"use_cache": true,
|
| 29 |
"use_sliding_window": false,
|
| 30 |
-
"vocab_size":
|
| 31 |
"summary_token_begin": 151936,
|
| 32 |
"summary_chunk_size": 8,
|
| 33 |
"summary_token_num": 1,
|
| 34 |
"use_summary_attention": true,
|
| 35 |
-
"summary_sliding_chunk_num": "([128]*3+[
|
|
|
|
| 36 |
"summary_chunk_position_ids_type": "origin",
|
| 37 |
"summary_token_position_ids_type": "last_chunk_slice_right",
|
| 38 |
"summary_independent_parameters": true,
|
| 39 |
"summary_independent_attention_layernorm": false,
|
| 40 |
-
"summary_attention_mode": "kernel",
|
| 41 |
"auto_map": {
|
| 42 |
-
"AutoModel": "
|
| 43 |
-
"AutoModelForCausalLM": "
|
| 44 |
}
|
| 45 |
}
|
| 46 |
-
|
|
|
|
| 8 |
"eos_token_id": 151643,
|
| 9 |
"head_dim": 128,
|
| 10 |
"hidden_act": "silu",
|
| 11 |
+
"hidden_size": 2560,
|
| 12 |
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 9728,
|
| 14 |
"max_position_embeddings": 131072,
|
| 15 |
+
"max_window_layers": 36,
|
| 16 |
+
"mix_coeff": 0,
|
| 17 |
"model_type": "qwen3",
|
| 18 |
+
"num_attention_heads": 32,
|
| 19 |
+
"num_hidden_layers": 36,
|
| 20 |
+
"num_key_value_heads": 8,
|
| 21 |
"rms_norm_eps": 1e-06,
|
| 22 |
"rope_scaling": null,
|
| 23 |
+
"rope_theta": 1000000,
|
| 24 |
"sliding_window": null,
|
| 25 |
"tie_word_embeddings": true,
|
| 26 |
"torch_dtype": "bfloat16",
|
| 27 |
"transformers_version": "4.51.0",
|
| 28 |
"use_cache": true,
|
| 29 |
"use_sliding_window": false,
|
| 30 |
+
"vocab_size": 151937,
|
| 31 |
"summary_token_begin": 151936,
|
| 32 |
"summary_chunk_size": 8,
|
| 33 |
"summary_token_num": 1,
|
| 34 |
"use_summary_attention": true,
|
| 35 |
+
"summary_sliding_chunk_num": "([128]*3+[16768]*1)*9",
|
| 36 |
+
"summary_layer_freq": "([1])*36",
|
| 37 |
"summary_chunk_position_ids_type": "origin",
|
| 38 |
"summary_token_position_ids_type": "last_chunk_slice_right",
|
| 39 |
"summary_independent_parameters": true,
|
| 40 |
"summary_independent_attention_layernorm": false,
|
|
|
|
| 41 |
"auto_map": {
|
| 42 |
+
"AutoModel": "modeling_qwen3.Qwen3Model",
|
| 43 |
+
"AutoModelForCausalLM": "modeling_qwen3.Qwen3ForCausalLM"
|
| 44 |
}
|
| 45 |
}
|
|
|
generation_config.json
CHANGED
|
@@ -1,7 +1,9 @@
|
|
| 1 |
{
|
|
|
|
| 2 |
"bos_token_id": 151643,
|
| 3 |
-
"do_sample": false,
|
| 4 |
"eos_token_id": 151643,
|
| 5 |
-
"
|
| 6 |
-
"
|
| 7 |
-
|
|
|
|
|
|
|
|
|
| 1 |
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
"bos_token_id": 151643,
|
|
|
|
| 4 |
"eos_token_id": 151643,
|
| 5 |
+
"output_attentions": false,
|
| 6 |
+
"output_hidden_states": false,
|
| 7 |
+
"transformers_version": "5.0.0",
|
| 8 |
+
"use_cache": true
|
| 9 |
+
}
|