Comments (5)
Could you please clarify if this discussion is around full parameter tuning or lora based? @bloc97
from yarn.
Thanks a lot. To confirm, A100 is with 40G or 80G memory for 7b 64k fine-tuning?
from yarn.
We were able to train the 7b 64k model on an 8x A100 node -- all other models unfortunately require a multinode setup. We used 64 GPUs, but I expect 16 would suffice for all other models (7b 128k, 13b 64k, 13b 128k)
from yarn.
It is 8x80GB for 64k context size
from yarn.
We were able to train the 7b 64k model on an 8x A100 node -- all other models unfortunately require a multinode setup. We used 64 GPUs, but I expect 16 would suffice for all other models (7b 128k, 13b 64k, 13b 128k)
I ran finetune.py using 2x A100 GPUs, and both GPUs loaded up to 14g/80g. After processing the first batch, the memory usage went up to 77g/80g, and then it ran OOM when starting the second batch.
Is this situation normal?
from yarn.
Related Issues (20)
- context length and dataset size
- Inquiry Regarding Evaluation Metrics in Your Paper HOT 2
- RoPE scaling config confusing
- Question about Yarn environment configuration (v2) HOT 4
- Running Error HOT 2
- deepspeed config crashed for `auto` and OOM HOT 3
- cannot load safetensor: Trying to set a tensor of shape torch.Size([0]) in "weight" (which has shape torch.Size([32000, 4096])) HOT 4
- Unexpected larger perplexity on PG19 HOT 1
- OOM on two 80GB GPUs HOT 6
- Could this repository be used for sft based on YaRN?
- Phi 2
- An OOM error occurred while computing the perplexity of 128k Proofpoint documents with a maximum token count set to 128k.
- Questions about DynamicNTK
- How should I proceed with conducting an evaluation for lm-evaluation-harness?
- Can we run the replication of the results,8 * 80 A100 HOT 1
- Trying to set a tensor of shape torch.Size([257, 1024]) in "weight" (which has shape torch.Size([1226, 1024])), this look incorrect
- Why the updated cache is initialized with seqlen=256?
- cannot connect to hugging face
- OOM error of distributed training on 80GB GPUs with Mistral-7b HOT 2
- Question related to _yarn_linear_ramp_mask 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 yarn.