vince62s commited on
Commit
b79d82a
1 Parent(s): 08fbb73

Upload 5 files

Browse files
Mistral-7B-instruct-onmt-awq-gemm.00.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:65bb17c50d659525fc657a8e36669452d9f2ad07c60f9998d3132d971ca40986
3
+ size 4150957864
Mistral-7B-instruct-onmt-awq-gemm.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f53362911e0907982fbae27a31545217d71bd25e576e9935a7ba6d3ad97ff46c
3
+ size 1054088
mistral-7b-awq-gemm-onmt-inference.yaml ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ transforms: [sentencepiece]
2
+
3
+ #### Subword
4
+ src_subword_model: "/mnt/InternalCrucial4/dataAI/mistral-7B/mistral-instruct-v0.2/tokenizer.model"
5
+ tgt_subword_model: "/mnt/InternalCrucial4/dataAI/mistral-7B/mistral-instruct-v0.2/tokenizer.model"
6
+
7
+ # Model info
8
+ model: "/mnt/InternalCrucial4/dataAI/mistral-7B/mistral-instruct-v0.2/Mistral-7B-instruct-onmt-awq-gemm.pt"
9
+
10
+ # Inference
11
+ seed: 42
12
+ max_length: 256
13
+ gpu: 0
14
+ batch_type: sents
15
+ batch_size: 60
16
+ world_size: 1
17
+ gpu_ranks: [0]
18
+ #parallel_mode: "tensor_parallel"
19
+ #quant_layers: ['w_1', 'w_2', 'w_3', 'linear_values', 'linear_query', 'linear_keys', 'final_linear']
20
+ #quant_type: "bnb_NF4"
21
+ precision: fp16
22
+ #random_sampling_topk: 1
23
+ #random_sampling_topp: 0.6
24
+ #random_sampling_temp: 0.9
25
+ beam_size: 1
26
+ n_best: 1
27
+ profile: false
28
+ report_time: true
29
+ src: None
30
+ #tgt: None
tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
3
+ size 493443
vocab.txt ADDED
The diff for this file is too large to render. See raw diff