koesn commited on
Commit
0151d58
1 Parent(s): f823ffe

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +47 -0
README.md CHANGED
@@ -1,3 +1,50 @@
1
  ---
2
  license: apache-2.0
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
  license: apache-2.0
3
  ---
4
+ # Garten2-7B-GGUF
5
+
6
+
7
+ ## Description
8
+
9
+ This repo contains GGUF format model files for Garten2-7B-GGUF.
10
+
11
+ ## Files Provided
12
+
13
+ | Name | Quant | Bits | File Size | Remark |
14
+ | ---------------------- | ------ | ---- | --------- | -------------------------------- |
15
+ | garten2-7b.IQ3_S.gguf | IQ3_S | 3 | 3.18 GB | 3.44 bpw quantization |
16
+ | garten2-7b.IQ3_M.gguf | IQ3_M | 3 | 3.28 GB | 3.66 bpw quantization mix |
17
+ | garten2-7b.Q4_0.gguf | Q4_0 | 4 | 4.11 GB | 3.56G, +0.2166 ppl |
18
+ | garten2-7b.IQ4_NL.gguf | IQ4_NL | 4 | 4.16 GB | 4.25 bpw non-linear quantization |
19
+ | garten2-7b.Q4_K_M.gguf | Q4_K_M | 4 | 4.37 GB | 3.80G, +0.0532 ppl |
20
+ | garten2-7b.Q5_K_M.gguf | Q5_K_M | 5 | 5.13 GB | 4.45G, +0.0122 ppl |
21
+ | garten2-7b.Q6_K.gguf | Q6_K | 6 | 5.94 GB | 5.15G, +0.0008 ppl |
22
+ | garten2-7b.Q8_0.gguf | Q8_0 | 8 | 7.70 GB | 6.70G, +0.0004 ppl |
23
+
24
+ ## Parameters
25
+
26
+ | path | type | architecture | rope_theta | sliding_win | max_pos_embed |
27
+ | -------------------- | ------- | ------------------ | ---------- | ----------- | ------------- |
28
+ | senseable/Garten2-7B | mistral | MistralForCausalLM | 10000.0 | 4096 | 32768 |
29
+
30
+ ## Benchmarks
31
+
32
+ ![](https://i.ibb.co/TmszRLV/Garten2-7-B.png)
33
+
34
+ # Original Model Card
35
+
36
+ ---
37
+ base_model:
38
+ - mistralai/Mistral-7B-v0.1
39
+ tags:
40
+ - qlora
41
+ - dto
42
+ language:
43
+ - "en"
44
+ library_name: transformers
45
+ license: "apache-2.0"
46
+ ---
47
+
48
+ # Details
49
+
50
+ Introducing Garten2-7B, a cutting-edge, small 7B all-purpose Language Model (LLM), designed to redefine the boundaries of artificial intelligence in natural language understanding and generation. Garten2-7B stands out with its unique architecture, expertly crafted to deliver exceptional performance in a wide array of tasks, from conversation to content creation.