Instructions to use SaurabhJ20/deva1-sft with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use SaurabhJ20/deva1-sft with Transformers:

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

pipe = pipeline("text-generation", model="SaurabhJ20/deva1-sft")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("SaurabhJ20/deva1-sft")
model = AutoModelForCausalLM.from_pretrained("SaurabhJ20/deva1-sft", device_map="auto")
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 SaurabhJ20/deva1-sft with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "SaurabhJ20/deva1-sft"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "SaurabhJ20/deva1-sft",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/SaurabhJ20/deva1-sft

SGLang

How to use SaurabhJ20/deva1-sft 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 "SaurabhJ20/deva1-sft" \
    --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": "SaurabhJ20/deva1-sft",
		"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 "SaurabhJ20/deva1-sft" \
        --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": "SaurabhJ20/deva1-sft",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use SaurabhJ20/deva1-sft with Docker Model Runner:
```
docker model run hf.co/SaurabhJ20/deva1-sft
```

See axolotl config

axolotl version: 0.13.0.dev0

base_model: SaurabhJ20/deva1-pretrain
tokenizer_type: AutoTokenizer
model_type: AutoModelForCausalLM
load_in_8bit: false
load_in_4bit: false
strict: false
datasets:
- path: pretraining_subset_3002.jsonl
  type: completion
- path: axolotl_correction_conversations_storeflow.json
  type: input_output
- path: axolotl_rag_conversations_storeflow.jsonl
  type: input_output
- path: factual_sft_completion/combined_all_0.jsonl
  type: completion
- path: factual_sft_completion/combined_all_1.jsonl
  type: completion
- path: generic_sft_completion/Augmentoolkit-Augmentoolkit-Pippa-Thoughts_200000.jsonl
  type: completion
- path: generic_sft_completion/Augmentoolkit-Augmentoolkit-Bluemoon-1mil-thoughts_200000.jsonl
  type: completion
- path: generic_sft_completion/Augmentoolkit-Augmentoolkit-Generic-Grabbag-Thoughts_400000.jsonl
  type: completion
- path: generic_sft_completion/Augmentoolkit-Openthoughts-100mil-DifferentFormat_800000.jsonl
  type: completion
- path: generic_sft_completion/Augmentoolkit-Augmentoolkit-LMsys-800k-Thoughts_200000.jsonl
  type: completion
- path: generic_sft_completion/Augmentoolkit-Augmentoolkit-Capybara-2point5mil-Thoughts_200000.jsonl
  type: completion
dataset_prepared_path: last_finetune_prepared
output_dir: ./finetune-model-output
seed: 1337
sequence_len: 5000
sample_packing: true
pad_to_sequence_len: false
shuffle_merged_datasets: true
gradient_accumulation_steps: 8
micro_batch_size: 1
eval_batch_size: 4
num_epochs: 5
optimizer: paged_adamw_8bit
lr_scheduler: constant
learning_rate: 2.0e-05
noisy_embedding_alpha: 5
weight_decay: 0
train_on_inputs: false
group_by_length: false
bf16: true
fp16: false
tf32: false
gradient_checkpointing: true
logging_steps: 1
xformers_attention: false
flash_attention: true
chat_template: chatml
auto_resume_from_checkpoints: false
warmup_ratio: 0.1
evals_per_epoch: 1
val_set_size: 0.04
saves_per_epoch: 1
eval_sample_packing: false
save_total_limit: 2
special_tokens:
  pad_token: <unk>
use_liger_kernel: true
plugins:
- axolotl.integrations.liger.LigerPlugin
liger_rope: true
liger_rms_norm: true
liger_glu_activation: true
liger_layer_norm: true
liger_fused_linear_cross_entropy: true
sequence_length: 10000
wandb_project: test-project
wandb_entity: ''
wandb_watch: ''
wandb_run_id: ''
wandb_log_model: ''
hub_model_id: SaurabhJ20/deva1-sft
hub_strategy: end

deva1-sft

This model is a fine-tuned version of SaurabhJ20/deva1-pretrain on the pretraining_subset_3002.jsonl, the axolotl_correction_conversations_storeflow.json, the axolotl_rag_conversations_storeflow.jsonl, the factual_sft_completion/combined_all_0.jsonl, the factual_sft_completion/combined_all_1.jsonl, the generic_sft_completion/Augmentoolkit-Augmentoolkit-Pippa-Thoughts_200000.jsonl, the generic_sft_completion/Augmentoolkit-Augmentoolkit-Bluemoon-1mil-thoughts_200000.jsonl, the generic_sft_completion/Augmentoolkit-Augmentoolkit-Generic-Grabbag-Thoughts_400000.jsonl, the generic_sft_completion/Augmentoolkit-Openthoughts-100mil-DifferentFormat_800000.jsonl, the generic_sft_completion/Augmentoolkit-Augmentoolkit-LMsys-800k-Thoughts_200000.jsonl and the generic_sft_completion/Augmentoolkit-Augmentoolkit-Capybara-2point5mil-Thoughts_200000.jsonl datasets. It achieves the following results on the evaluation set:

Loss: 1.2796
Memory/max Mem Active(gib): 29.42
Memory/max Mem Allocated(gib): 29.42
Memory/device Mem Reserved(gib): 31.53

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: 1
eval_batch_size: 4
seed: 1337
gradient_accumulation_steps: 8
total_train_batch_size: 8
optimizer: Use OptimizerNames.PAGED_ADAMW_8BIT with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: constant
lr_scheduler_warmup_steps: 29
training_steps: 290

Training results

Training Loss	Epoch	Step	Validation Loss	Mem Active(gib)	Mem Allocated(gib)	Mem Reserved(gib)
No log	0	0	1.6836	19.63	19.63	26.05
0.9692	0.9872	58	0.9428	29.42	29.42	31.46
0.6315	1.9702	116	0.9980	29.42	29.42	31.46
0.2773	2.9532	174	1.1260	29.42	29.42	31.51
0.216	3.9362	232	1.2337	29.42	29.42	31.51
0.0997	4.9191	290	1.2796	29.42	29.42	31.53

Framework versions

Transformers 4.55.0
Pytorch 2.7.1+cu128
Datasets 4.0.0
Tokenizers 0.21.4

Downloads last month: 7

Safetensors

Model size

7B params

Tensor type

BF16

Model tree for SaurabhJ20/deva1-sft

Base model

mistral-community/Mistral-7B-v0.2

Finetuned

SaurabhJ20/deva1-pretrain

Finetuned

(1)

this model