Skip to content

issues Search Results · repo:Lightning-AI/litgpt language:Python

Filter by

895 results
 (83 ms)

895 results

inLightning-AI/litgpt (press backspace or delete to remove)

Bug description i modified the generate/base.py , one inference with kvcache, and the other one is without kvcache, i set both temperature=0, topk=None and topp=0, i use the same seed, i use the same ...
bug
  • guangyuli-uoe
  • 3
  • Opened 
    2 days ago
  • #2165

Cut cross entropy has been shown in my own work to be really easy to use and saves a lot of memory when training models with large vocab sizes. Therefore I think adding it to litGPT would help people fit ...
enhancement
  • Aceticia
  • Opened 
    9 days ago
  • #2163

Hi, Was wondering if adding support for SSM models (Mamba, Mamba2) would be something you d consider having in litgpt? https://arxiv.org/abs/2405.21060 Thanks!
enhancement
  • radulescupetru
  • Opened 
    15 days ago
  • #2158

Bug description When I use litgpt chat, I am observing that it converts the prompt I provide to all lowercase before feeding it to the model I am chatting with. The line of code where this happens is ...
bug
  • drwslacy47
  • 3
  • Opened 
    23 days ago
  • #2154

litgpt==0.5.11 litdata==0.2.58, OS: windows 10, python==3.13, torch==2.9.0+cu126 I m following the tutorials to pretrain a smollm2-135M with tinystories dataset, and my script is like this: import lightning ...
question
  • li-yang23
  • 2
  • Opened 
    on Oct 30
  • #2151

jsonargparse==4.42.0 is causing test_config_help failure while it passes untill 4.41.0 (litgpt) ⚡ fix/tests_config_hub ~/litgpt uv pip install jsonargparse==4.41.0 (litgpt) ⚡ fix/tests_config_hub ~/litgpt ...
  • bhimrazy
  • Opened 
    on Oct 30
  • #2150

Bug description from litgpt import LLM llm = LLM.load( google/gemma-3-4b-it ) # KeyError: transformer.wte.weight litgpt download list shows that this model should be usable. google/gemma-3-1b-it does ...
bug
  • nsfinkelstein
  • 4
  • Opened 
    on Oct 27
  • #2147

hi, i would like to use different learning rate for different layer, currently im using the pretrain.py, but i have no idea how to do it in litgpt, looking forward to your kind reply,
question
  • guangyuli-uoe
  • 4
  • Opened 
    on Oct 20
  • #2146

Bug description I’m using litgpt (Version: 0.5.11) to chat with the following models: checkpoints/google/gemma-3-4b-it checkpoints/meta-llama/Llama-3.2-3B-Instruct I deliberately prompt both models ...
bug
  • drwslacy47
  • Opened 
    on Oct 19
  • #2145

Bug description I am using litgpt 0.5.11 and litdata 0.2.58 (installed following instructions from the pretrain tutorial) and I am getting the following error message when running the pretraining demo. ...
bug
  • th789
  • 1
  • Opened 
    on Oct 12
  • #2144
Issue origami icon

Learn how you can use GitHub Issues to plan and track your work.

Save views for sprints, backlogs, teams, or releases. Rank, sort, and filter issues to suit the occasion. The possibilities are endless.Learn more about GitHub Issues
ProTip! 
Restrict your search to the title by using the in:title qualifier.
Issue origami icon

Learn how you can use GitHub Issues to plan and track your work.

Save views for sprints, backlogs, teams, or releases. Rank, sort, and filter issues to suit the occasion. The possibilities are endless.Learn more about GitHub Issues
ProTip! 
Restrict your search to the title by using the in:title qualifier.
Issue search results · GitHub