Upload LlamaForCausalLM
Browse files- config.json +14 -2
- pytorch_model.bin +3 -0
config.json
CHANGED
@@ -1,5 +1,5 @@
|
|
1 |
{
|
2 |
-
"_name_or_path": "
|
3 |
"architectures": [
|
4 |
"LlamaForCausalLM"
|
5 |
],
|
@@ -15,11 +15,23 @@
|
|
15 |
"num_hidden_layers": 32,
|
16 |
"num_key_value_heads": 32,
|
17 |
"pretraining_tp": 1,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
18 |
"rms_norm_eps": 1e-05,
|
19 |
"rope_scaling": null,
|
20 |
"rope_theta": 1000000,
|
21 |
"tie_word_embeddings": false,
|
22 |
-
"torch_dtype": "
|
23 |
"transformers_version": "4.33.1",
|
24 |
"use_cache": true,
|
25 |
"vocab_size": 32016
|
|
|
1 |
{
|
2 |
+
"_name_or_path": "merged_model",
|
3 |
"architectures": [
|
4 |
"LlamaForCausalLM"
|
5 |
],
|
|
|
15 |
"num_hidden_layers": 32,
|
16 |
"num_key_value_heads": 32,
|
17 |
"pretraining_tp": 1,
|
18 |
+
"quantization_config": {
|
19 |
+
"bnb_4bit_compute_dtype": "float32",
|
20 |
+
"bnb_4bit_quant_type": "fp4",
|
21 |
+
"bnb_4bit_use_double_quant": false,
|
22 |
+
"llm_int8_enable_fp32_cpu_offload": false,
|
23 |
+
"llm_int8_has_fp16_weight": false,
|
24 |
+
"llm_int8_skip_modules": null,
|
25 |
+
"llm_int8_threshold": 6.0,
|
26 |
+
"load_in_4bit": false,
|
27 |
+
"load_in_8bit": true,
|
28 |
+
"quant_method": "bitsandbytes"
|
29 |
+
},
|
30 |
"rms_norm_eps": 1e-05,
|
31 |
"rope_scaling": null,
|
32 |
"rope_theta": 1000000,
|
33 |
"tie_word_embeddings": false,
|
34 |
+
"torch_dtype": "float16",
|
35 |
"transformers_version": "4.33.1",
|
36 |
"use_cache": true,
|
37 |
"vocab_size": 32016
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4d0d66e56fe81adf899cf9617f55651fda45db9e170891c899158a832c76d203
|
3 |
+
size 7006701357
|