Comments (12)
Hi guys, I just tested the 30B model, it works fine (with the conversion manually from https://gist.github.com/eiz/828bddec6162a023114ce19146cb2b82) and don't forget the modification in the llama cpp to load a single file for the 30b model.
from serge.
I think you don't need to grab a different tokenizer, I believe they're exactly the same.
You can check it here:
https://huggingface.co/decapoda-research/llama-30b-hf/blob/main/tokenizer.model
https://huggingface.co/decapoda-research/llama-13b-hf/blob/main/tokenizer.model
https://huggingface.co/decapoda-research/llama-7b-hf/blob/main/tokenizer.model
They all have the same SHA256 hash
from serge.
Yes you right no need, i used same
from serge.
I've made the 30B model available to download but I don't have the hardware to test it. So if someone feels like it, feel free to download it ! Instructions are in the README.
from serge.
I can test it, but it looks like the 7B tokenizer is downloaded. I'm running main at ref 58cf7d0
from serge.
I hacked it locally with this, but it's pretty jank. I think the model should determine the tokenizer.
index 73298b5..d0eafcb 100644
--- a/api/utils/download.py
+++ b/api/utils/download.py
@@ -10,6 +10,7 @@ models_info = {
"13B": ["Pi3141/alpaca-13B-ggml", "ggml-model-q4_0.bin"],
"30B": ["Pi3141/alpaca-30B-ggml", "ggml-model-q4_0.bin"],
"tokenizer": ["decapoda-research/llama-7b-hf", "tokenizer.model"],
+ "30B-tokenizer": ["decapoda-research/llama-30b-hf", "tokenizer.model"],
}
@@ -21,7 +22,7 @@ def parse_args():
"model",
help="Model name",
nargs="+",
- choices=["7B", "13B", "30B", "tokenizer"],
+ choices=["7B", "13B", "30B", "tokenizer", "30B-tokenizer"],
)
return parser.parse_args()
from serge.
Hi guys, I just tested the 30B model, it works fine (with the conversion manually from https://gist.github.com/eiz/828bddec6162a023114ce19146cb2b82) and don't forget the modification in the llama cpp to load a single file for the 30b model.
Are you sure this is needed ? I was pretty sure the --n_parts
argument to llama.cpp could let you handle it without modifying the source.
In serge it's handled there:
https://github.com/nsarrazin/serge/blob/58cf7d0f451035843d8b7e8cac048513fe47e01c/api/utils/generate.py#L47-L48
from serge.
Sorry, you're right, I was on the old method in my head, I had forgotten the features in command line.
from serge.
I hacked it locally with this, but it's pretty jank. I think the model should determine the tokenizer.
index 73298b5..d0eafcb 100644 --- a/api/utils/download.py +++ b/api/utils/download.py @@ -10,6 +10,7 @@ models_info = { "13B": ["Pi3141/alpaca-13B-ggml", "ggml-model-q4_0.bin"], "30B": ["Pi3141/alpaca-30B-ggml", "ggml-model-q4_0.bin"], "tokenizer": ["decapoda-research/llama-7b-hf", "tokenizer.model"], + "30B-tokenizer": ["decapoda-research/llama-30b-hf", "tokenizer.model"], } @@ -21,7 +22,7 @@ def parse_args(): "model", help="Model name", nargs="+", - choices=["7B", "13B", "30B", "tokenizer"], + choices=["7B", "13B", "30B", "tokenizer", "30B-tokenizer"], ) return parser.parse_args()
Thanks for doing this do you know if it's actually necessary to grab the matching tokenizer.model
?
The instructions here ggerganov/llama.cpp#382 (comment) just mention grabbing a tokenizer, and so I assumed you could use the tokenizer from the 7B repo for all the weights. I'm gonna test for myself it that works still.
And are you able to get any outputs from the 30B model with Serge so far ? @dacamp
from serge.
Thanks @maximeseth
from serge.
Seems like this could be closed then ?
from serge.
I'm closing this, I think it works. If it doesn't I'll reopen it.
from serge.
Related Issues (20)
- 🚀 [Feature]: Add OpenVino / OpenVino Model Server HOT 1
- 🐛 [Bug]: Web interface does not render properly on mobile devices HOT 1
- 🚀 [Feature]: Add LINCE-Mistal model HOT 1
- 🐛 [Bug]: UI components are missing accessibility labels HOT 2
- 🐛 [Bug]: response text generated by a model sometimes disappears after computer/browser is woken up from a 'sleep' HOT 4
- have a separate page which displays downloaded moddles. HOT 1
- 🚀 [Feature]: Add support for Intel ARC GPUs A750 and A770 (If Possible) HOT 2
- bug: Allow loading .gguf and .bin files HOT 3
- 🚀 [Feature]: add eagle 7b HOT 3
- 🐛 [Bug]: system reachable via ICMP and via Port 8008 but screen "navy blue" with no text whatsoever HOT 14
- 🚀 [Feature]: Add Nous-Hermes-2-Mistral-7B-DPO HOT 8
- 🚀 [Feature]: Add support for uploading files during chat conversation
- 🐛 [Bug]: New install - response keeps repeating the last line HOT 7
- 🚀 [Feature]: add characters HOT 6
- 🚀 [Feature]: Please add Gorilla: Large Language Model Connected with Massive APIs HOT 3
- 🤗 [Question]: Whats the difference between the... models?
- 🚀 [Feature]: Add meta-llama/Meta-Llama-3-70B-Instruct HOT 7
- 🐛 [Bug]: Can't use pre-existing model at /weights HOT 1
- 🐛 [Bug]: DLLAMA_BLAS_VENDOR=OpenBLAS build with pip is not enabling OpenBlas HOT 3
- how to use mixtral-8x7b-v0.1🤗 [Question]: HOT 2
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from serge.