Instructions to use Ba2han/kk-lora-t with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Ba2han/kk-lora-t with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("Ba2han/kk-lora-t", dtype="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps
- Unsloth Studio new
How to use Ba2han/kk-lora-t with Unsloth Studio:
Install Unsloth Studio (macOS, Linux, WSL)
curl -fsSL https://unsloth.ai/install.sh | sh # Run unsloth studio unsloth studio -H 0.0.0.0 -p 8888 # Then open http://localhost:8888 in your browser # Search for Ba2han/kk-lora-t to start chatting
Install Unsloth Studio (Windows)
irm https://unsloth.ai/install.ps1 | iex # Run unsloth studio unsloth studio -H 0.0.0.0 -p 8888 # Then open http://localhost:8888 in your browser # Search for Ba2han/kk-lora-t to start chatting
Using HuggingFace Spaces for Unsloth
# No setup required # Open https://huggingface.co/spaces/unsloth/studio in your browser # Search for Ba2han/kk-lora-t to start chatting
Load model with FastModel
pip install unsloth from unsloth import FastModel model, tokenizer = FastModel.from_pretrained( model_name="Ba2han/kk-lora-t", max_seq_length=2048, )
Training in progress, step 538
Browse files- README.md +59 -0
- adapter_config.json +50 -0
- adapter_model.safetensors +3 -0
- chat_template.jinja +5 -0
- tokenizer.json +0 -0
- tokenizer_config.json +12 -0
- training_args.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,59 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: AlicanKiraz0/Kara-Kumru-v1.0-2B
|
| 3 |
+
library_name: transformers
|
| 4 |
+
model_name: kk-lora-t
|
| 5 |
+
tags:
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
- sft
|
| 8 |
+
- trl
|
| 9 |
+
- unsloth
|
| 10 |
+
licence: license
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model Card for kk-lora-t
|
| 14 |
+
|
| 15 |
+
This model is a fine-tuned version of [AlicanKiraz0/Kara-Kumru-v1.0-2B](https://huggingface.co/AlicanKiraz0/Kara-Kumru-v1.0-2B).
|
| 16 |
+
It has been trained using [TRL](https://github.com/huggingface/trl).
|
| 17 |
+
|
| 18 |
+
## Quick start
|
| 19 |
+
|
| 20 |
+
```python
|
| 21 |
+
from transformers import pipeline
|
| 22 |
+
|
| 23 |
+
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
|
| 24 |
+
generator = pipeline("text-generation", model="Ba2han/kk-lora-t", device="cuda")
|
| 25 |
+
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
|
| 26 |
+
print(output["generated_text"])
|
| 27 |
+
```
|
| 28 |
+
|
| 29 |
+
## Training procedure
|
| 30 |
+
|
| 31 |
+
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/batuhan409/huggingface/runs/c285xupl)
|
| 32 |
+
|
| 33 |
+
|
| 34 |
+
|
| 35 |
+
This model was trained with SFT.
|
| 36 |
+
|
| 37 |
+
### Framework versions
|
| 38 |
+
|
| 39 |
+
- TRL: 1.0.0.dev0
|
| 40 |
+
- Transformers: 5.3.0
|
| 41 |
+
- Pytorch: 2.8.0
|
| 42 |
+
- Datasets: 4.3.0
|
| 43 |
+
- Tokenizers: 0.22.2
|
| 44 |
+
|
| 45 |
+
## Citations
|
| 46 |
+
|
| 47 |
+
|
| 48 |
+
|
| 49 |
+
Cite TRL as:
|
| 50 |
+
|
| 51 |
+
```bibtex
|
| 52 |
+
@software{vonwerra2020trl,
|
| 53 |
+
title = {{TRL: Transformers Reinforcement Learning}},
|
| 54 |
+
author = {von Werra, Leandro and Belkada, Younes and Tunstall, Lewis and Beeching, Edward and Thrush, Tristan and Lambert, Nathan and Huang, Shengyi and Rasul, Kashif and Gallouédec, Quentin},
|
| 55 |
+
license = {Apache-2.0},
|
| 56 |
+
url = {https://github.com/huggingface/trl},
|
| 57 |
+
year = {2020}
|
| 58 |
+
}
|
| 59 |
+
```
|
adapter_config.json
ADDED
|
@@ -0,0 +1,50 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alora_invocation_tokens": null,
|
| 3 |
+
"alpha_pattern": {},
|
| 4 |
+
"arrow_config": null,
|
| 5 |
+
"auto_mapping": {
|
| 6 |
+
"base_model_class": "MistralForCausalLM",
|
| 7 |
+
"parent_library": "transformers.models.mistral.modeling_mistral",
|
| 8 |
+
"unsloth_fixed": true
|
| 9 |
+
},
|
| 10 |
+
"base_model_name_or_path": "AlicanKiraz0/Kara-Kumru-v1.0-2B",
|
| 11 |
+
"bias": "lora_only",
|
| 12 |
+
"corda_config": null,
|
| 13 |
+
"ensure_weight_tying": false,
|
| 14 |
+
"eva_config": null,
|
| 15 |
+
"exclude_modules": null,
|
| 16 |
+
"fan_in_fan_out": false,
|
| 17 |
+
"inference_mode": true,
|
| 18 |
+
"init_lora_weights": true,
|
| 19 |
+
"layer_replication": null,
|
| 20 |
+
"layers_pattern": null,
|
| 21 |
+
"layers_to_transform": null,
|
| 22 |
+
"loftq_config": {},
|
| 23 |
+
"lora_alpha": 64,
|
| 24 |
+
"lora_bias": false,
|
| 25 |
+
"lora_dropout": 0,
|
| 26 |
+
"megatron_config": null,
|
| 27 |
+
"megatron_core": "megatron.core",
|
| 28 |
+
"modules_to_save": null,
|
| 29 |
+
"peft_type": "LORA",
|
| 30 |
+
"peft_version": "0.18.1",
|
| 31 |
+
"qalora_group_size": 16,
|
| 32 |
+
"r": 64,
|
| 33 |
+
"rank_pattern": {},
|
| 34 |
+
"revision": null,
|
| 35 |
+
"target_modules": [
|
| 36 |
+
"gate_proj",
|
| 37 |
+
"up_proj",
|
| 38 |
+
"o_proj",
|
| 39 |
+
"down_proj",
|
| 40 |
+
"q_proj",
|
| 41 |
+
"v_proj",
|
| 42 |
+
"k_proj"
|
| 43 |
+
],
|
| 44 |
+
"target_parameters": null,
|
| 45 |
+
"task_type": "CAUSAL_LM",
|
| 46 |
+
"trainable_token_indices": null,
|
| 47 |
+
"use_dora": true,
|
| 48 |
+
"use_qalora": false,
|
| 49 |
+
"use_rslora": false
|
| 50 |
+
}
|
adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b0eca7a674f5563af1fc8e415f82a796178253e705b17b89335df99384fb8f6d
|
| 3 |
+
size 273582264
|
chat_template.jinja
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>
|
| 2 |
+
|
| 3 |
+
'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|>
|
| 4 |
+
|
| 5 |
+
' }}{% endif %}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"backend": "tokenizers",
|
| 3 |
+
"bos_token": "<BOS>",
|
| 4 |
+
"clean_up_tokenization_spaces": true,
|
| 5 |
+
"eos_token": "<EOS>",
|
| 6 |
+
"is_local": false,
|
| 7 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 8 |
+
"pad_token": "<PAD>",
|
| 9 |
+
"padding_side": "right",
|
| 10 |
+
"tokenizer_class": "TokenizersBackend",
|
| 11 |
+
"unk_token": "<UNK>"
|
| 12 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3110b0ee1563e4353f21914876215bdd33492eb40396ec0def9d6a01b3c9c01d
|
| 3 |
+
size 5777
|