issues
search
unslothai
/
unsloth
Finetune Llama 3.2, Mistral, Phi, Qwen 2.5 & Gemma LLMs 2-5x faster with 80% less memory
https://unsloth.ai
Apache License 2.0
18.41k
stars
1.29k
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Unsloth: Most labels in your dataset are -100. Training losses will be 0.
#1128
Hasan-Demez
closed
1 month ago
2
Can i fine tune models that are not on unsloth/<model_name> list
#1127
miloskovacevic68
closed
1 month ago
2
The issue Multiple dispatch failed for 'torch._ops.aten.to.dtype_layout' is relevant again
#1126
yurkoff-mv
opened
1 month ago
3
Counting untrained tokens ? What is it doing ? Took forever on large dataset, and repeating !
#1125
thusinh1969
opened
1 month ago
1
Unable to load locally located adapter
#1124
yurkoff-mv
closed
1 month ago
2
Continued pretraining facing catastrophic forgetting
#1123
InderjeetVishnoi
closed
1 month ago
3
Is it possible for the merge16bit to fail sometimes?
#1122
brando90
opened
1 month ago
9
Only remove folder in sentencepiece check if it was created
#1121
giuliabaldini
closed
1 month ago
1
Handle absolute paths for save_to_gguf using pathlib
#1120
giuliabaldini
closed
1 month ago
1
Model folder deleted after saving
#1119
giuliabaldini
closed
1 month ago
4
save_to_gguf doesn't work with absolute paths
#1118
giuliabaldini
closed
1 month ago
2
colab: with BentoML
#1117
aarnphm
closed
1 month ago
3
Compatibility issues with CUDA 12.4
#1116
seetharamarao817
opened
1 month ago
1
Time Overhead Comparison: unsloth RoPE vs. transformers Llama Rotary Embedding
#1115
xlim1996
closed
1 month ago
1
I decodes the prediction samples of Llama but it seems not right, am I training it well?
#1114
diazr04
opened
1 month ago
2
Getting error while deploying the GGUF to ollama
#1113
InderjeetVishnoi
closed
1 month ago
1
How to save a llama3.2 model?
#1112
fzyzcjy
closed
1 month ago
2
Training has become slower since Oct 7 2024.
#1111
mahiatlinux
opened
1 month ago
4
ValueError: You cannot perform fine-tuning on purely quantized models. Please attach trainable adapters on top of the quantized model to correctly perform fine-tuning.
#1110
s0ul141
closed
1 month ago
0
`_fast_inner_training_loop` has ZeroDivisionError
#1109
fzyzcjy
closed
1 month ago
1
Resize embeddings, tokenizers - adding new tokens don't work
#1108
danielhanchen
opened
1 month ago
3
Error loading model saved to file
#1107
laura-burdick-sil
closed
1 month ago
3
Can I Fine-Tune a Model on CPU Using Unsloth?
#1106
OE-LUCIFER
opened
1 month ago
1
Tied weights like Llama 3.2 3B cannot save during checkpointing
#1105
kovern
opened
1 month ago
9
Unable to run saving GGUF F16, KeyError: '"name"'.
#1104
ramzyizza
opened
1 month ago
5
feature request: vision model support (Qwen2-VL 7b)
#1103
charlesmindee
closed
23 hours ago
3
An error while NOT using the train_on_response only
#1102
Yashar78
closed
1 month ago
2
Getting CUDA OOM on training gemma-2-2b with "lm_head" and "embed_token" target projects.
#1101
InderjeetVishnoi
opened
1 month ago
6
Query regarding deployment of unsloth trained models.
#1100
InderjeetVishnoi
opened
1 month ago
6
NotImplementedError: Make sure that a `_reorder_cache` function is correctly implemented in transformers.models.llama.modeling_llama to enable beam search for <class 'transformers.models.llama.modeling_llama.LlamaForCausalLM'>
#1099
kiranpedvak
opened
1 month ago
2
colab: add example colab with BentoML
#1098
aarnphm
closed
1 month ago
2
Problem with finetuning Qwen 2.5 model
#1097
Pstva
closed
1 month ago
1
Adding a custom model
#1096
HARISHSENTHIL
closed
1 month ago
2
Nvidia L4 inference speed
#1095
OsaCode
closed
1 month ago
5
AttributeError: 'LlamaForCausalLM' object has no attribute 'save_pretrained_gguf'
#1094
awesomecoolraj
opened
1 month ago
3
Lora adapter is almost as large as model
#1093
kirawi
opened
1 month ago
5
adding new language/continuous pretraining notebook
#1092
Pranil51
opened
1 month ago
2
ValueError: Invalid `cache_implementation` (dynamic). Choose one of: ['static', 'offloaded_static', 'sliding_window', 'hybrid', 'mamba', 'quantized', 'static']
#1091
AzinY
closed
1 month ago
3
Are there any guidelines for loading a CPT (continued pre-training) model and retraining it on a different data set?
#1090
daegonYu
closed
2 weeks ago
9
Unsloth QLora merging into base-model: what is the best practice if you want to run trained model with vLLM or NVIDIA TensorRT-LLM ?
#1089
thusinh1969
opened
1 month ago
3
NotImplementedError: Unsloth: unsloth/Llama-3.2-11B-Vision-Instruct-bnb-4bit not supported yet!
#1088
one-and-only
closed
23 hours ago
6
Keyname "name" GGUF saving
#1087
marscuspolos
closed
3 weeks ago
5
chore(build): move cli within package
#1086
SauravMaheshkar
opened
1 month ago
2
Error in introducing task_type as TOKEN_CLS
#1085
yaswanthan
opened
1 month ago
1
FastLanguageModel: AttributeError: module 'pyarrow.lib' has no attribute 'ListViewType'
#1084
tapankumarpatro
opened
1 month ago
1
Update README.md
#1083
danielhanchen
closed
1 month ago
0
Introduce MsT technologies into unsloth to extend sequence length
#1082
wdlctc
opened
1 month ago
9
Mini-Sequence Transformer integration
#1081
Trapper4888
opened
1 month ago
1
Bitsandbytes issue
#1080
StrangeTcy
opened
1 month ago
6
Fix merges
#1079
danielhanchen
closed
1 month ago
0
Previous
Next