Upload README.md with huggingface_hub
Browse files
README.md
ADDED
@@ -0,0 +1,90 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
base_model: BEE-spoke-data/smol_llama-220M-GQA
|
3 |
+
datasets:
|
4 |
+
- JeanKaddour/minipile
|
5 |
+
- pszemraj/simple_wikipedia_LM
|
6 |
+
- mattymchen/refinedweb-3m
|
7 |
+
- BEE-spoke-data/knowledge-inoc-concat-v1
|
8 |
+
inference: false
|
9 |
+
language:
|
10 |
+
- en
|
11 |
+
license: apache-2.0
|
12 |
+
model_creator: BEE-spoke-data
|
13 |
+
model_name: smol_llama-220M-GQA
|
14 |
+
pipeline_tag: text-generation
|
15 |
+
quantized_by: afrideva
|
16 |
+
tags:
|
17 |
+
- smol_llama
|
18 |
+
- llama2
|
19 |
+
- gguf
|
20 |
+
- ggml
|
21 |
+
- quantized
|
22 |
+
- q2_k
|
23 |
+
- q3_k_m
|
24 |
+
- q4_k_m
|
25 |
+
- q5_k_m
|
26 |
+
- q6_k
|
27 |
+
- q8_0
|
28 |
+
widget:
|
29 |
+
- example_title: El Microondas
|
30 |
+
text: My name is El Microondas the Wise, and
|
31 |
+
- example_title: Kennesaw State University
|
32 |
+
text: Kennesaw State University is a public
|
33 |
+
- example_title: Bungie
|
34 |
+
text: Bungie Studios is an American video game developer. They are most famous for
|
35 |
+
developing the award winning Halo series of video games. They also made Destiny.
|
36 |
+
The studio was founded
|
37 |
+
- example_title: Mona Lisa
|
38 |
+
text: The Mona Lisa is a world-renowned painting created by
|
39 |
+
- example_title: Harry Potter Series
|
40 |
+
text: The Harry Potter series, written by J.K. Rowling, begins with the book titled
|
41 |
+
- example_title: Riddle
|
42 |
+
text: 'Question: I have cities, but no houses. I have mountains, but no trees. I
|
43 |
+
have water, but no fish. What am I?
|
44 |
+
|
45 |
+
Answer:'
|
46 |
+
- example_title: Photosynthesis
|
47 |
+
text: The process of photosynthesis involves the conversion of
|
48 |
+
- example_title: Story Continuation
|
49 |
+
text: Jane went to the store to buy some groceries. She picked up apples, oranges,
|
50 |
+
and a loaf of bread. When she got home, she realized she forgot
|
51 |
+
- example_title: Math Problem
|
52 |
+
text: 'Problem 2: If a train leaves Station A at 9:00 AM and travels at 60 mph,
|
53 |
+
and another train leaves Station B at 10:00 AM and travels at 80 mph, when will
|
54 |
+
they meet if the distance between the stations is 300 miles?
|
55 |
+
|
56 |
+
To determine'
|
57 |
+
- example_title: Algorithm Definition
|
58 |
+
text: In the context of computer programming, an algorithm is
|
59 |
+
---
|
60 |
+
# BEE-spoke-data/smol_llama-220M-GQA-GGUF
|
61 |
+
|
62 |
+
Quantized GGUF model files for [smol_llama-220M-GQA](https://huggingface.co/BEE-spoke-data/smol_llama-220M-GQA) from [BEE-spoke-data](https://huggingface.co/BEE-spoke-data)
|
63 |
+
|
64 |
+
|
65 |
+
| Name | Quant method | Size |
|
66 |
+
| ---- | ---- | ---- |
|
67 |
+
| [smol_llama-220m-gqa.fp16.gguf](https://huggingface.co/afrideva/smol_llama-220M-GQA-GGUF/resolve/main/smol_llama-220m-gqa.fp16.gguf) | fp16 | 436.50 MB |
|
68 |
+
| [smol_llama-220m-gqa.q2_k.gguf](https://huggingface.co/afrideva/smol_llama-220M-GQA-GGUF/resolve/main/smol_llama-220m-gqa.q2_k.gguf) | q2_k | 102.60 MB |
|
69 |
+
| [smol_llama-220m-gqa.q3_k_m.gguf](https://huggingface.co/afrideva/smol_llama-220M-GQA-GGUF/resolve/main/smol_llama-220m-gqa.q3_k_m.gguf) | q3_k_m | 115.70 MB |
|
70 |
+
| [smol_llama-220m-gqa.q4_k_m.gguf](https://huggingface.co/afrideva/smol_llama-220M-GQA-GGUF/resolve/main/smol_llama-220m-gqa.q4_k_m.gguf) | q4_k_m | 137.58 MB |
|
71 |
+
| [smol_llama-220m-gqa.q5_k_m.gguf](https://huggingface.co/afrideva/smol_llama-220M-GQA-GGUF/resolve/main/smol_llama-220m-gqa.q5_k_m.gguf) | q5_k_m | 157.91 MB |
|
72 |
+
| [smol_llama-220m-gqa.q6_k.gguf](https://huggingface.co/afrideva/smol_llama-220M-GQA-GGUF/resolve/main/smol_llama-220m-gqa.q6_k.gguf) | q6_k | 179.52 MB |
|
73 |
+
| [smol_llama-220m-gqa.q8_0.gguf](https://huggingface.co/afrideva/smol_llama-220M-GQA-GGUF/resolve/main/smol_llama-220m-gqa.q8_0.gguf) | q8_0 | 232.28 MB |
|
74 |
+
|
75 |
+
|
76 |
+
|
77 |
+
## Original Model Card:
|
78 |
+
# smol_llama: 220M GQA
|
79 |
+
|
80 |
+
> model card WIP, more details to come
|
81 |
+
|
82 |
+
|
83 |
+
A small 220M param (total) decoder model. This is the first version of the model.
|
84 |
+
|
85 |
+
- 1024 hidden size, 10 layers
|
86 |
+
- GQA (32 heads, 8 key-value), context length 2048
|
87 |
+
- train-from-scratch on one GPU :)
|
88 |
+
|
89 |
+
|
90 |
+
---
|