Instructions to use CaffeineThief/ttp_sft_kanana-1.5_routing_dataset_3epoch with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use CaffeineThief/ttp_sft_kanana-1.5_routing_dataset_3epoch with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="CaffeineThief/ttp_sft_kanana-1.5_routing_dataset_3epoch")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("CaffeineThief/ttp_sft_kanana-1.5_routing_dataset_3epoch")
model = AutoModelForCausalLM.from_pretrained("CaffeineThief/ttp_sft_kanana-1.5_routing_dataset_3epoch")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use CaffeineThief/ttp_sft_kanana-1.5_routing_dataset_3epoch with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "CaffeineThief/ttp_sft_kanana-1.5_routing_dataset_3epoch"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "CaffeineThief/ttp_sft_kanana-1.5_routing_dataset_3epoch",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/CaffeineThief/ttp_sft_kanana-1.5_routing_dataset_3epoch

SGLang

How to use CaffeineThief/ttp_sft_kanana-1.5_routing_dataset_3epoch with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "CaffeineThief/ttp_sft_kanana-1.5_routing_dataset_3epoch" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "CaffeineThief/ttp_sft_kanana-1.5_routing_dataset_3epoch",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "CaffeineThief/ttp_sft_kanana-1.5_routing_dataset_3epoch" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "CaffeineThief/ttp_sft_kanana-1.5_routing_dataset_3epoch",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use CaffeineThief/ttp_sft_kanana-1.5_routing_dataset_3epoch with Docker Model Runner:
```
docker model run hf.co/CaffeineThief/ttp_sft_kanana-1.5_routing_dataset_3epoch
```

ttp_sft_kanana-1.5_routing_dataset_3epoch

File size: 3,120 Bytes

c531142

---
library_name: transformers
license: apache-2.0
base_model: kakaocorp/kanana-1.5-2.1b-instruct-2505
tags:
- axolotl
- generated_from_trainer
datasets:
- datasets/routing_dataset.jsonl
model-index:
- name: ttp_sft_kanana-1.5_routing_dataset_3epoch
  results: []
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

[<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
<details><summary>See axolotl config</summary>

axolotl version: `0.12.2`
```yaml
base_model: kakaocorp/kanana-1.5-2.1b-instruct-2505
hf_cache_dir: ../../../../data5/models

load_in_8bit: false
load_in_4bit: false

datasets:
  - path: datasets/routing_dataset.jsonl
    type: chat_template
    split: train

dataset_prepared_path: preprocess
val_set_size: 0
output_dir: ./outputs-llm-router-3epoch
dataloader_num_workers: 32

sequence_len: 3072
sample_packing: false
eval_sample_packing: false
pad_to_sequence_len: false

plugins:
  - axolotl.integrations.liger.LigerPlugin
liger_rope: true
liger_rms_norm: true
liger_swiglu: true
liger_fused_linear_cross_entropy: true

wandb_project: TTP_SFT_LLM_RE
wandb_entity:
wandb_watch:
wandb_name: ttp_sft_kanana-1.5_routing_dataset_3epoch
wandb_log_model:
hub_model_id: CaffeineThief/ttp_sft_kanana-1.5_routing_dataset_3epoch
hub_private_repo: false

gradient_accumulation_steps: 1
micro_batch_size: 16
num_epochs: 3
optimizer: adamw_torch_fused
lr_scheduler: cosine
learning_rate: 2e-5

bf16: auto
tf32: false

gradient_checkpointing: false
resume_from_checkpoint:
logging_steps: 1
flash_attention: true

warmup_ratio: 0.05
weight_decay: 0.01
evals_per_epoch: 1
saves_per_epoch: 1

fsdp:
  - full_shard
  - auto_wrap
fsdp_config:
  fsdp_state_dict_type: FULL_STATE_DICT
  fsdp_transformer_layer_cls_to_wrap: LlamaDecoderLayer
  fsdp_activation_checkpointing: true
```

</details><br>

# ttp_sft_kanana-1.5_routing_dataset_3epoch

This model is a fine-tuned version of [kakaocorp/kanana-1.5-2.1b-instruct-2505](https://huggingface.co/kakaocorp/kanana-1.5-2.1b-instruct-2505) on the datasets/routing_dataset.jsonl dataset.

## Model description

More information needed

## Intended uses & limitations

More information needed

## Training and evaluation data

More information needed

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- distributed_type: multi-GPU
- num_devices: 3
- total_train_batch_size: 48
- total_eval_batch_size: 48
- optimizer: Use adamw_torch_fused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 14
- training_steps: 282

### Training results



### Framework versions

- Transformers 4.55.2
- Pytorch 2.6.0+cu124
- Datasets 4.0.0
- Tokenizers 0.21.4