Comments (6)
Embedding models are going to be supported in v0.4.3
So, install from source to use now
from vllm.
Thank you very much. I will try to install new version.
from vllm.
Hello, I would like to ask why this error:
"---------------------------------------------------------------------------
AttributeError Traceback (most recent call last)
in <cell line: 1>()
----> 1 outputs = model.encode(prompts)
10 frames
/content/vllm/vllm/model_executor/sampling_metadata.py in _prepare_seq_groups(seq_group_metadata_list, seq_lens, query_lens, device)
206
207 if seq_group_metadata.is_prompt:
--> 208 if sampling_params.seed is not None:
209 seq_group_metadata.state.generator = torch.Generator(
210 device=device).manual_seed(sampling_params.seed)
AttributeError: 'NoneType' object has no attribute 'seed'"occurs during use?
The model I'm using is llama3-8b, thanks!
from vllm.
Hello, I would like to ask why this error: "--------------------------------------------------------------------------- AttributeError Traceback (most recent call last) in <cell line: 1>() ----> 1 outputs = model.encode(prompts)
10 frames /content/vllm/vllm/model_executor/sampling_metadata.py in _prepare_seq_groups(seq_group_metadata_list, seq_lens, query_lens, device) 206 207 if seq_group_metadata.is_prompt: --> 208 if sampling_params.seed is not None: 209 seq_group_metadata.state.generator = torch.Generator( 210 device=device).manual_seed(sampling_params.seed) AttributeError: 'NoneType' object has no attribute 'seed'"occurs during use?
The model I'm using is llama3-8b, thanks!
If you're running the generic Llama-3-8b model, you are running with a generation model rather than an embedding model. I should update the encode
API to fail more gracefully if called in this configuration.
If you want to use an embedding model, try:
from vllm.
Hello, I would like to ask why this error: "--------------------------------------------------------------------------- AttributeError Traceback (most recent call last) in <cell line: 1>() ----> 1 outputs = model.encode(prompts)
10 frames /content/vllm/vllm/model_executor/sampling_metadata.py in _prepare_seq_groups(seq_group_metadata_list, seq_lens, query_lens, device) 206 207 if seq_group_metadata.is_prompt: --> 208 if sampling_params.seed is not None: 209 seq_group_metadata.state.generator = torch.Generator( 210 device=device).manual_seed(sampling_params.seed) AttributeError: 'NoneType' object has no attribute 'seed'"occurs during use?
The model I'm using is llama3-8b, thanks!If you're running the generic Llama-3-8b model, you are running with a generation model rather than an embedding model. I should update the
encode
API to fail more gracefully if called in this configuration.If you want to use an embedding model, try:
Thank you very much for your quick reply! May I ask if there is an API for vllm if I want to extract the hidden state of the generation model?
from vllm.
Hello, I would like to ask why this error: "--------------------------------------------------------------------------- AttributeError Traceback (most recent call last) in <cell line: 1>() ----> 1 outputs = model.encode(prompts)
10 frames /content/vllm/vllm/model_executor/sampling_metadata.py in _prepare_seq_groups(seq_group_metadata_list, seq_lens, query_lens, device) 206 207 if seq_group_metadata.is_prompt: --> 208 if sampling_params.seed is not None: 209 seq_group_metadata.state.generator = torch.Generator( 210 device=device).manual_seed(sampling_params.seed) AttributeError: 'NoneType' object has no attribute 'seed'"occurs during use?
The model I'm using is llama3-8b, thanks!If you're running the generic Llama-3-8b model, you are running with a generation model rather than an embedding model. I should update the
encode
API to fail more gracefully if called in this configuration.
If you want to use an embedding model, try:Thank you very much for your quick reply! May I ask if there is an API for vllm if I want to extract the hidden state of the generation model?
Getting the hidden state of the generation model is exactly what I am struggling with. Have you found any good solutions? All the tools I found are targeted for embedding model, not supporting llama3.
from vllm.
Related Issues (20)
- [Installation]: vLLM source install on rocm 6.2 still requires libamdhip64.so.6 HOT 4
- [Performance]: Issues with prefix cache usage
- [Performance]: INT4 quantisation does not lead to any observable throughput increase HOT 3
- [Bug]: vLLM with Neuron performance degrades dramatically if request concurrency is >= max_num_seqs
- [Misc]/[Tracking]: FP8 Datatype parameter for Flashinfer backend Metadata accumulation for its decode wrapper.
- [Doc]: nvidia ammo has been renamed HOT 1
- [Bug]: v0.5.5 crash: "AssertionError: expected running sequences" HOT 26
- [Bug]: InternVL2-2B outputs gibberish with tensor parallel inference
- [Bug]: 批量任务报错
- [Bug]: Trailing newline as outputs HOT 1
- [Usage]: Can and How we start server on multi-node multi-gpu with torchrun? HOT 4
- [New Model]: Want to support BAAI/bge-reranker-v2-m3 model
- [Bug]: Flashinfer now supports SM75, but VLLM is still encountering errors.
- [Bug]: deploy on V100, mma -> mma layout conversion is only supported on Ampere HOT 2
- [Bug]: RuntimeError: CUDA error: an illegal memory access was encountered HOT 4
- [Bug]: I get one word inconsistent responses using the v0.5.4 with LL HOT 1
- [Installation]: LGPL license in dependencies HOT 8
- [Usage]: How can I determine the maximum number of concurrent requests? HOT 1
- [Misc]: Question about Serving with Server API HOT 7
- [Feature]: Contribute T5 model to vLLM HOT 1
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 vllm.