Upload folder using huggingface_hub
Browse files
flcc.bpe
ADDED
The diff for this file is too large to render.
See raw diff
|
|
flcc.json
ADDED
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_flash_attn_2_enabled": true,
|
3 |
+
"architectures": [
|
4 |
+
"LlamaForCausalLM"
|
5 |
+
],
|
6 |
+
"attention_bias": false,
|
7 |
+
"attention_dropout": 0.0,
|
8 |
+
"bos_token_id": 1,
|
9 |
+
"eos_token_id": 2,
|
10 |
+
"hidden_act": "silu",
|
11 |
+
"hidden_size": 1024,
|
12 |
+
"initializer_range": 0.02,
|
13 |
+
"intermediate_size": 2816,
|
14 |
+
"max_position_embeddings": 1536,
|
15 |
+
"model_type": "llama",
|
16 |
+
"num_attention_heads": 8,
|
17 |
+
"num_hidden_layers": 6,
|
18 |
+
"num_key_value_heads": 8,
|
19 |
+
"pretraining_tp": 1,
|
20 |
+
"rms_norm_eps": 1e-06,
|
21 |
+
"rope_scaling": null,
|
22 |
+
"rope_theta": 10000.0,
|
23 |
+
"tie_word_embeddings": false,
|
24 |
+
"torch_dtype": "float32",
|
25 |
+
"transformers_version": "4.36.0",
|
26 |
+
"use_cache": true,
|
27 |
+
"vocab_size": 16384
|
28 |
+
}
|
flcc.model
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5fb5cbd759167c3be670688ed5147aba153187ca2a003dcdafcb19258264454b
|
3 |
+
size 68636512
|
model.xml
ADDED
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
<!--AUTO-GENERATED FILE. DO NOT MODIFY. xml file is generated by generateModel gradle task-->
|
2 |
+
<model>
|
3 |
+
<version>0.1.5-native-llama-bundle</version>
|
4 |
+
<size>68855513</size>
|
5 |
+
<jar>local-model-php-0.1.5-native-llama-bundle.jar</jar>
|
6 |
+
<languages>
|
7 |
+
<language>php</language>
|
8 |
+
</languages>
|
9 |
+
<changelog>
|
10 |
+
- Init LLaMA model for PHP
|
11 |
+
</changelog>
|
12 |
+
<binary>flcc.model</binary>
|
13 |
+
<bpe>flcc.bpe</bpe>
|
14 |
+
<config>flcc.json</config>
|
15 |
+
<inferenceType>LLAMA_NATIVE</inferenceType>
|
16 |
+
<native>
|
17 |
+
<archive>full-line-inference.zip</archive>
|
18 |
+
<version>1.0.84</version>
|
19 |
+
</native>
|
20 |
+
</model>
|