Comments (4)
I also tried Opt 1.3B following this tutorial, and I'm getting awful output.
[
"</s>Hello, I am a all I the,ly):,) A for\n in and) a,, A: and to: to to Ay,.\n on, To and in, A, ]\n.\n A\n,). A,,,*\n?..",
"</s>When did the Queen die robot the�, by- A I in in, A of,\n. that to\n:,, is To to�, A is., is A A to A A A that A, A,:s,), the A: to to A\n can A,."
]
from transformers-neuronx.
Same as @harryjulian on inf2.8xlarge using Deep Learning Base Neuron Ubuntu 20.04 AMI 20230612 - ami-0b453554dc1a73411
['</s>The quick brown fox and and and a big.\n\n" \'cause of the the they may it took to the same same is being.\nThe the city.\n1,“\n4-10.\n"\n6__ _ _ _ _\nIn In Allowed\n\n-\n-</s>']
from transformers-neuronx.
We are able to reproduce it and are root-causing the issue. Will update the issue once we have the fix.
from transformers-neuronx.
With release 2.12, the OPT result now look normal:
(aws_neuron_venv_pytorch) ubuntu@ip-10-0-10-149:~$ python opt.py
/home/ubuntu/aws_neuron_venv_pytorch/lib/python3.8/site-packages/transformers_neuronx/opt/model.py:37: UserWarning: torch_dtype=torch.float32 ignored in favor of amp=f32
warnings.warn(f'torch_dtype={config.torch_dtype} ignored in favor of amp={amp}')
..
Compiler status PASS
..
Compiler status PASS
..
Compiler status PASS
..
Compiler status PASS
..
Compiler status PASS
["</s>The quick brown fox jumps over the lazy dog. I just can't stop laughing at the image of a fox jumping over a lazy dog...\nI don't see what's funny about it. It's just a fox jumping over a lazy dog.</s>"]
Another output of another run:
['</s>The quick brown fox jumps over the lazy dog.\nThe lazy dog jumps over the quick brown fox. FTFY</s>']
Neuron packages used:
(aws_neuron_venv_pytorch) ubuntu@ip-10-0-10-149:~$ pip list | grep neuron
aws-neuronx-runtime-discovery 2.9
libneuronxla 0.5.391
neuronx-cc 2.8.0.25+a3ad0f342
neuronx-distributed 0.1.0
neuronx-hwm 2.8.0.3+2b7c6da39
torch-neuronx 1.13.1.1.9.0
torch-xla 1.13.1+torchneuron8
transformers-neuronx 0.5.58
from transformers-neuronx.
Related Issues (20)
- How to use generate() with inputs_embeds HOT 2
- Mixtral config issue -- not handling null well HOT 8
- Generate Llama 2 from Embeddings HOT 5
- Infering logits from `model.forward` for the entire batch instead of the last forward's output. HOT 6
- Support for MPT model HOT 1
- `stopping_criteria_list(input_ids, probs)` does not check for the correct sequence. HOT 4
- User feedback when compiling and reloading a large model HOT 1
- Issue while compiling Mistral 7B 0.2 Instruct HOT 5
- Backward compatibility with saved llama 2 compiled artifacts HOT 1
- NaN outputs when masking llama model inputs HOT 8
- Improve Neuron model loading time HOT 4
- Add support for `gemma` models HOT 1
- Add support for Baichuan-13B model
- Latest changes introduced for continuous batching break Mixtral model HOT 5
- llava support HOT 3
- Any plan to support Qwen-2 Model
- Neuron model NEFFs are dependent on the python path HOT 2
- Not able to load llama 3 70b on inf2.24xlarge instance HOT 5
- Gibberish output for princeton-nlp/Sheared-LLaMA-1.3B with continuous batching HOT 2
- [Question] BasicTransformerBlock
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 transformers-neuronx.