Giters
johnsmith0031
/
alpaca_lora_4bit
Geek Repo:
Geek Repo
Github PK Tool:
Github PK Tool
Stargazers:
533
Watchers:
12
Issues:
113
Forks:
86
johnsmith0031/alpaca_lora_4bit Issues
Support for moe model?
Updated
4 months ago
Comments count
2
Why lora support is only for simple lora with only q_proj and v_proj ?
Updated
5 months ago
Comments count
1
trying to get this working with text-generation-webui
Updated
9 months ago
Comments count
9
Error attempting to finetune llama2-70b
Updated
10 months ago
Comments count
5
AttributeError: 'dict' object has no attribute 'to_dict'
Closed
a year ago
Comments count
1
Finetuning CodeLLaMA34B - RuntimeError: The size of tensor a (1024) must match the size of tensor b (8192)
Closed
a year ago
Comments count
3
docker.io/nvidia/cuda:11.7.0-devel-ubuntu22.04 not available anymore
Closed
a year ago
Comments count
1
3 errors detected in the compilation of "src/alpaca_lora_4bit/quant_cuda/quant_cuda_kernel.cu"
Closed
a year ago
Comments count
2
monkeypatch problem
Updated
a year ago
Comments count
8
ValueError: Target module Autograd4bitQuantLinear() is not supported.
Closed
a year ago
Comments count
7
Target module Autograd4bitQuantLinear() is not supported
Updated
a year ago
Comments count
5
OOM on inference while i can finetune with more tokens
Closed
a year ago
Comments count
2
module 'alpaca_lora_4bit.quant_cuda' has no attribute 'vecquant4recons_v2'
Closed
a year ago
Comments count
4
Unable to Build Wheels
Closed
a year ago
Comments count
8
Merging LoRA after finetune
Updated
a year ago
Comments count
1
Targeting all layers and biases
Closed
a year ago
Comments count
2
Checkpoint saving broken with the latest version of huggingface
Closed
a year ago
Comments count
8
Feature request: Stop when loss reaches X
Updated
a year ago
Comments count
1
Is alpaca_lora_4bit@winglian-setup_pip missing finetune.py?
Updated
a year ago
Comments count
1
High perplexity while lower loss after LoRA finetuning (how?)
Closed
a year ago
Comments count
5
LoRA Output Identical to Base Model
Closed
a year ago
Comments count
4
Flash Attention 2
Closed
a year ago
Comments count
1
How to use inference.py after finetune.py?
Closed
a year ago
Comments count
2
TypeError: object of type 'NoneType' has no len()
Closed
a year ago
Comments count
1
Gibberish results for non-disabled "faster_mode" using "vicuna-7B-GPTQ-4bit-128g" model
Updated
a year ago
Comments count
4
July
Closed
a year ago
Comments count
4
Crashes during finetuning
Updated
a year ago
Comments count
2
Update docs for > 2048 token models (SuperHOT)?
Updated
a year ago
Comments count
12
Differences between QLoRA and this repo
Updated
a year ago
Comments count
2
Inf or NaN in probabilities. Windows 10, vicuna-7b-gptq-4bit-128g
Closed
a year ago
Comments count
35
this repo support 2bit finetuning the llama model? Is there any case to show how to run the scripts?
Updated
a year ago
Comments count
1
[question] weights in the replaced quantized modules
Closed
a year ago
how to change into 8 bit
Updated
a year ago
Comments count
1
Problem with inference
Closed
a year ago
Comments count
7
fine tune with 2 GPU
Updated
a year ago
Comments count
2
Version of GPTQ
Updated
a year ago
Comments count
2
how to infer with finetuned model?
Updated
a year ago
Comments count
4
ImportError: cannot import name '_get_submodules' from 'peft.utils'
Closed
a year ago
Comments count
10
Consider using new QLoRA
Updated
a year ago
Comments count
3
Implementing Landmark Attention
Updated
a year ago
Finetuning 2-bit Quantized Models
Updated
a year ago
Comments count
7
When running with 2 GPUs, got multiple values for keyword argument 'backend'
Closed
a year ago
Comments count
20
Code reference request
Updated
a year ago
Comments count
1
Problem loading safetensor file format
Updated
a year ago
Comments count
3
what is the difference between v1 model and v2 model?
Updated
a year ago
Comments count
1
ValueError: Autograd4bitQuantLinear() does not have a parameter or a buffer named qzeros.
Updated
a year ago
Comments count
2
error with monkeypatch and model gpt-j and lora
Updated
a year ago
TypeError: '<' not supported between instances of 'tuple' and 'float' while trying to generate completion through the v2 13bit LLAMA
Closed
a year ago
Comments count
6
Which script were used for 4bit quantization?
Updated
a year ago
Comments count
2
run_server.sh: ValueError: Autograd4bitQuantLinear() does not have a parameter or a buffer named g_idx.
Updated
a year ago
Comments count
1
Previous
Next