Adding 15M, 42M & 110M models
Browse files- README.md +21 -15
- stories110Mtok32000.gguf +3 -0
- stories15Mtok4096.gguf +3 -0
- stories42Mtok32000.gguf +3 -0
- stories42Mtok4096.gguf +3 -0
README.md
CHANGED
@@ -12,7 +12,18 @@ language:
|
|
12 |
- The models were created with the training procedure outlined in [karpathy/llama2.c](https://github.com/karpathy/llama2.c)
|
13 |
- You can run them local too, as described in [karpathy/llama2.c](https://github.com/karpathy/llama2.c)
|
14 |
|
15 |
-
##
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
16 |
|
17 |
See: [Getting Started: set-up](https://huggingface.co/docs/hub/repositories-getting-started#set-up)
|
18 |
|
@@ -22,26 +33,20 @@ pip install huggingface-hub
|
|
22 |
git clone <this-repo>
|
23 |
cd <this-repo>
|
24 |
|
|
|
25 |
git lfs install
|
26 |
-
git lfs track "*.gguf"
|
27 |
huggingface-cli lfs-enable-largefiles .
|
28 |
|
29 |
-
#
|
|
|
|
|
|
|
30 |
git add <file-name>
|
31 |
git commit -m "Adding <file-name>"
|
32 |
git push -u origin main
|
33 |
```
|
34 |
|
35 |
-
##
|
36 |
-
|
37 |
-
| model | notes |
|
38 |
-
|-------|-------|
|
39 |
-
| stories260Ktok512.guff | Use this for development & debugging |
|
40 |
-
| stories15Mtok4096.guff | Fits in canister & works well ! |
|
41 |
-
| stories42Mtok4096.guff | As of April 28, hits instruction limit of canister |
|
42 |
-
| stories42Mtok32000.guff (*) | As of April 28, hits instruction limit of canister |
|
43 |
-
| stories110Mtok32000.guff (*) | As of April 28, hits instruction limit of canister |
|
44 |
-
|
45 |
|
46 |
We used [convert-llama2c-to-ggml](https://github.com/ggerganov/llama.cpp/tree/32c8486e1f0297393cb22ac0a0d26a6b17ad4d54/examples/convert-llama2c-to-ggml) to convert the llama2.c model+tokenizer to llama.cpp gguf format.
|
47 |
- Good read: [lama : add support for llama2.c models](https://github.com/ggerganov/llama.cpp/issues/2379)
|
@@ -64,8 +69,9 @@ convert-llama2c-to-ggml --llama2c-model stories42Mtok32000.bin --copy-vocab-from
|
|
64 |
main -m stories15Mtok4096.gguf -p "Joe loves writing stories" -n 600 -c 128
|
65 |
|
66 |
# Quantization
|
67 |
-
#
|
68 |
```
|
69 |
|
70 |
-
|
|
|
71 |
|
|
|
12 |
- The models were created with the training procedure outlined in [karpathy/llama2.c](https://github.com/karpathy/llama2.c)
|
13 |
- You can run them local too, as described in [karpathy/llama2.c](https://github.com/karpathy/llama2.c)
|
14 |
|
15 |
+
## TinyStories models
|
16 |
+
|
17 |
+
| model | notes |
|
18 |
+
|-------|-------|
|
19 |
+
| stories260Ktok512.guff | Use this for development & debugging |
|
20 |
+
| stories15Mtok4096.guff | Fits in canister & works well ! |
|
21 |
+
| stories42Mtok4096.guff | As of April 28, hits instruction limit of canister |
|
22 |
+
| stories42Mtok32000.guff | As of April 28, hits instruction limit of canister |
|
23 |
+
| stories110Mtok32000.guff | As of April 28, hits instruction limit of canister |
|
24 |
+
|
25 |
+
|
26 |
+
## Setup local git with lfs
|
27 |
|
28 |
See: [Getting Started: set-up](https://huggingface.co/docs/hub/repositories-getting-started#set-up)
|
29 |
|
|
|
33 |
git clone <this-repo>
|
34 |
cd <this-repo>
|
35 |
|
36 |
+
# configure lfs for local repo
|
37 |
git lfs install
|
|
|
38 |
huggingface-cli lfs-enable-largefiles .
|
39 |
|
40 |
+
# tell lfs what files to track (.gitattributes)
|
41 |
+
git lfs track "*.gguf"
|
42 |
+
|
43 |
+
# add, commit & push as usual with git
|
44 |
git add <file-name>
|
45 |
git commit -m "Adding <file-name>"
|
46 |
git push -u origin main
|
47 |
```
|
48 |
|
49 |
+
## Model creation
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
50 |
|
51 |
We used [convert-llama2c-to-ggml](https://github.com/ggerganov/llama.cpp/tree/32c8486e1f0297393cb22ac0a0d26a6b17ad4d54/examples/convert-llama2c-to-ggml) to convert the llama2.c model+tokenizer to llama.cpp gguf format.
|
52 |
- Good read: [lama : add support for llama2.c models](https://github.com/ggerganov/llama.cpp/issues/2379)
|
|
|
69 |
main -m stories15Mtok4096.gguf -p "Joe loves writing stories" -n 600 -c 128
|
70 |
|
71 |
# Quantization
|
72 |
+
# TODO
|
73 |
```
|
74 |
|
75 |
+
|
76 |
+
|
77 |
|
stories110Mtok32000.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1e92613ae001f20c05b690a7cb5044058b6f605457c0f0bbb8b5f61e9cc62630
|
3 |
+
size 537153728
|
stories15Mtok4096.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e1d6eb48fa458b38268d484bb667c8b7b25d72f5befda5ea58d7c6d6c9cf7da2
|
3 |
+
size 33436768
|
stories42Mtok32000.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c2d4c26604cebc4d76aaa83874af9cb3b012145494c9d9b28c2ba7d1b7952cf9
|
3 |
+
size 233022592
|
stories42Mtok4096.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b6957cb15eaa94dc7b5d4b6a4badbcd1acb8bd5c7e79bd6907924f2e1995845e
|
3 |
+
size 118097408
|