Hugging Face's logo

flozi00
/
open_llama_7b_v2-german-assistant

Text Generation
Transformers
PyTorch
English
German
llama
text-generation-inference
Inference Endpoints
Model card Files Community
flozi00 commited on
Commit
6446825
•
1 Parent(s): 0f43e07

Upload model

Browse files
Files changed (3) hide show
  1. README.md +116 -0
  2. adapter_config.json +26 -0
  3. adapter_model.bin +3 -0
README.md ADDED
@@ -0,0 +1,116 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+
7
+ The following `bitsandbytes` quantization config was used during training:
8
+ - load_in_8bit: False
9
+ - load_in_4bit: True
10
+ - llm_int8_threshold: 6.0
11
+ - llm_int8_skip_modules: None
12
+ - llm_int8_enable_fp32_cpu_offload: False
13
+ - llm_int8_has_fp16_weight: False
14
+ - bnb_4bit_quant_type: fp4
15
+ - bnb_4bit_use_double_quant: True
16
+ - bnb_4bit_compute_dtype: float16
17
+
18
+ The following `bitsandbytes` quantization config was used during training:
19
+ - load_in_8bit: False
20
+ - load_in_4bit: True
21
+ - llm_int8_threshold: 6.0
22
+ - llm_int8_skip_modules: None
23
+ - llm_int8_enable_fp32_cpu_offload: False
24
+ - llm_int8_has_fp16_weight: False
25
+ - bnb_4bit_quant_type: fp4
26
+ - bnb_4bit_use_double_quant: True
27
+ - bnb_4bit_compute_dtype: float16
28
+
29
+ The following `bitsandbytes` quantization config was used during training:
30
+ - load_in_8bit: False
31
+ - load_in_4bit: True
32
+ - llm_int8_threshold: 6.0
33
+ - llm_int8_skip_modules: None
34
+ - llm_int8_enable_fp32_cpu_offload: False
35
+ - llm_int8_has_fp16_weight: False
36
+ - bnb_4bit_quant_type: fp4
37
+ - bnb_4bit_use_double_quant: True
38
+ - bnb_4bit_compute_dtype: float16
39
+
40
+ The following `bitsandbytes` quantization config was used during training:
41
+ - load_in_8bit: False
42
+ - load_in_4bit: True
43
+ - llm_int8_threshold: 6.0
44
+ - llm_int8_skip_modules: None
45
+ - llm_int8_enable_fp32_cpu_offload: False
46
+ - llm_int8_has_fp16_weight: False
47
+ - bnb_4bit_quant_type: fp4
48
+ - bnb_4bit_use_double_quant: True
49
+ - bnb_4bit_compute_dtype: float16
50
+
51
+ The following `bitsandbytes` quantization config was used during training:
52
+ - load_in_8bit: False
53
+ - load_in_4bit: True
54
+ - llm_int8_threshold: 6.0
55
+ - llm_int8_skip_modules: None
56
+ - llm_int8_enable_fp32_cpu_offload: False
57
+ - llm_int8_has_fp16_weight: False
58
+ - bnb_4bit_quant_type: fp4
59
+ - bnb_4bit_use_double_quant: True
60
+ - bnb_4bit_compute_dtype: float16
61
+
62
+ The following `bitsandbytes` quantization config was used during training:
63
+ - load_in_8bit: False
64
+ - load_in_4bit: True
65
+ - llm_int8_threshold: 6.0
66
+ - llm_int8_skip_modules: None
67
+ - llm_int8_enable_fp32_cpu_offload: False
68
+ - llm_int8_has_fp16_weight: False
69
+ - bnb_4bit_quant_type: fp4
70
+ - bnb_4bit_use_double_quant: True
71
+ - bnb_4bit_compute_dtype: float16
72
+
73
+ The following `bitsandbytes` quantization config was used during training:
74
+ - load_in_8bit: False
75
+ - load_in_4bit: True
76
+ - llm_int8_threshold: 6.0
77
+ - llm_int8_skip_modules: None
78
+ - llm_int8_enable_fp32_cpu_offload: False
79
+ - llm_int8_has_fp16_weight: False
80
+ - bnb_4bit_quant_type: fp4
81
+ - bnb_4bit_use_double_quant: True
82
+ - bnb_4bit_compute_dtype: float16
83
+
84
+ The following `bitsandbytes` quantization config was used during training:
85
+ - load_in_8bit: False
86
+ - load_in_4bit: True
87
+ - llm_int8_threshold: 6.0
88
+ - llm_int8_skip_modules: None
89
+ - llm_int8_enable_fp32_cpu_offload: False
90
+ - llm_int8_has_fp16_weight: False
91
+ - bnb_4bit_quant_type: fp4
92
+ - bnb_4bit_use_double_quant: True
93
+ - bnb_4bit_compute_dtype: float16
94
+
95
+ The following `bitsandbytes` quantization config was used during training:
96
+ - load_in_8bit: False
97
+ - load_in_4bit: True
98
+ - llm_int8_threshold: 6.0
99
+ - llm_int8_skip_modules: None
100
+ - llm_int8_enable_fp32_cpu_offload: False
101
+ - llm_int8_has_fp16_weight: False
102
+ - bnb_4bit_quant_type: fp4
103
+ - bnb_4bit_use_double_quant: True
104
+ - bnb_4bit_compute_dtype: float16
105
+ ### Framework versions
106
+
107
+ - PEFT 0.4.0.dev0
108
+ - PEFT 0.4.0.dev0
109
+ - PEFT 0.4.0.dev0
110
+ - PEFT 0.4.0.dev0
111
+ - PEFT 0.4.0.dev0
112
+ - PEFT 0.4.0.dev0
113
+ - PEFT 0.4.0.dev0
114
+ - PEFT 0.4.0.dev0
115
+
116
+ - PEFT 0.4.0.dev0
adapter_config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "base_model_name_or_path": "openlm-research/open_llama_7b_v2",
3
+ "bias": "none",
4
+ "fan_in_fan_out": false,
5
+ "inference_mode": true,
6
+ "init_lora_weights": true,
7
+ "layers_pattern": null,
8
+ "layers_to_transform": null,
9
+ "lora_alpha": 64,
10
+ "lora_dropout": 0.0,
11
+ "modules_to_save": null,
12
+ "peft_type": "LORA",
13
+ "r": 32,
14
+ "revision": null,
15
+ "target_modules": [
16
+ "gate_proj",
17
+ "down_proj",
18
+ "up_proj",
19
+ "v_proj",
20
+ "q_proj",
21
+ "k_proj",
22
+ "o_proj",
23
+ "embed_tokens"
24
+ ],
25
+ "task_type": "CAUSAL_LM"
26
+ }
adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c81ba92493358f133b6044ba37b6d4fcad5bcfdb7daad34f0ddfc91561e887f8
3
+ size 324598229