-
### Feature request
Besides loss, users often need to report additional metrics throughout the training in order to drive decision making and communicate results, which in the case of Seq2Seq models …
-
Hello developers,
I am trying optimizers in python-ace, and I met a problem when using trf/lm/dogbox optimizer. Here is the error information when I set optimizer as 'lm' for the example of Cu-I:
``…
-
There are many llm api calls during pipeline optimization.
Does DSPY support asynchronous calls to LLMs yet? When trainset size up to hundreds level, the current sequential execution is too time-con…
Hspix updated
3 weeks ago
-
Were the effects of using different embedding models considered or tested? I'm just curious about it.
Related: https://github.com/lm-sys/RouteLLM/issues/23 https://github.com/lm-sys/RouteLLM/issues…
-
When I install the environment, it says that the package mlx is not found. Is this package private? How should I download it?
-
Is there a paper/article/blog post explaining such decision? Or is it just simply a feature that not being supported at the moment?
-
windows 2022 Datacenter
64gig ram
16 core
version 0.2.23
It ran successfully for me on three machines but this machine was not happy after upgrade and llama3 download.
the modal just does not see…
-
Using TabbyAPI/exllamav2 with Llama3.1-8B
Threadripper Pro/A6000 GPU
Inference at ~70t/s unconstrained, single request. ~35t/s with lm-format-enforcer (JSON schema)
Running 30 simultaneous reques…
-
I find [https://github.com/lm-sys/FastChat/pull/3423/commits/45cf5ccce897f5c86c812d3c729f2b9ac8e144b0](https://github.com/lm-sys/FastChat/pull/3423/commits/45cf5ccce897f5c86c812d3c729f2b9ac8e144b0)
…
-
### Create 4 new work types on LM-License Agreement
- Streetscape
- Tower Crane
- Retention System
- Other
**Notes:**
- We are only adding work types. No other functionality will change to t…