-
## training_commands:
```bash
openrlhf.cli.train_dpo
--save_path ./checkpoint/llama3-8b-dpo-lora
--save_steps -1
--logging_steps 1
--eval_steps -1
--train_batch_size 24
…
-
os: WINDOWS 10
File "D:\unsloth\unsloth\kernels\rms_layernorm.py", line 144, in forward
fx[(n_rows,)](
File "D:\ProgramData\anaconda3\envs\cuda\Lib\site-packages\triton\runtime\jit.py", l…
-
Has anyone been able to achieve this successfully ?
I've tried saving my model in gguf format with:
```
model.save_pretrained_gguf("model_gguf", tokenizer)
```
and then to convert it in onn…
-
### Why does Transformer Lens only support quantized LLaMA models?
Hi everyone,
I'm trying to use the transformer_lens library to study the activations of a quantized Mistral 7B model (unsloth/m…
-
We have a number of our cluster users trying out unsloth. Because we run LTS kernel versions, the kernel version check built into unsloth produces confusion. The kernel LTS versions are numerically mu…
-
Hello, thank you very much for your work. …
-
os: windows
I think my environment is ready
use jupyter notebook locally
when i run these:
"from unsloth import FastLanguageModel
import torch
max_seq_length = 8192 # Choose any! We auto sup…
-
I am now trying to finetune a llama3 model. I am using unsloth,
`from unsloth import FastLanguageModel`
Then I load Llama3 model.
```
model, tokenizer = FastLanguageModel.from_pretrained(
…
-
I'm trying to delete my conda env that I created during #98
```sh
❯ conda env list
# conda environments:
#
base /gpfs/share/rhel9/opt/spack/opt/spack/linux-rhel9-skylake_av…
-
Just a quick (and important) question about LoRA vs QLoRA with Unsloth.
I have read through a series of articles about DO NOT MERGED naievely QLoRA back to base model, it will give worse performanc…