Instructions to use Jboadu/GAIA-sft-v.03 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Jboadu/GAIA-sft-v.03 with Transformers:

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

pipe = pipeline("text-generation", model="Jboadu/GAIA-sft-v.03")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Jboadu/GAIA-sft-v.03")
model = AutoModelForCausalLM.from_pretrained("Jboadu/GAIA-sft-v.03")
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 Jboadu/GAIA-sft-v.03 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Jboadu/GAIA-sft-v.03"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Jboadu/GAIA-sft-v.03",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Jboadu/GAIA-sft-v.03

SGLang

How to use Jboadu/GAIA-sft-v.03 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 "Jboadu/GAIA-sft-v.03" \
    --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": "Jboadu/GAIA-sft-v.03",
		"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 "Jboadu/GAIA-sft-v.03" \
        --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": "Jboadu/GAIA-sft-v.03",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Jboadu/GAIA-sft-v.03 with Docker Model Runner:
```
docker model run hf.co/Jboadu/GAIA-sft-v.03
```

GAIA-sft-v.03 / README.md

Jboadu

End of training

0d4e329 verified 11 months ago

preview code

Raw

History Blame Contribute Delete

5.95 kB

metadata

library_name: transformers
license: apache-2.0
base_model: Jboadu/test-model-1-pretrain
tags:
  - axolotl
  - generated_from_trainer
datasets:
  - pretraining_subset_17828.jsonl
  - com_factual/allToolCallRejections.jsonl
  - com_factual/clean_output_0.jsonl
  - com_factual/clean_output_1.jsonl
  - com_factual/clean_output_2.jsonl
  - com_factual/clean_output_3.jsonl
  - com_factual/clean_output_4.jsonl
  - com_factual/clean_output_5.jsonl
  - com_factual/clean_output_6.jsonl
  - com_factual/clean_output_7.jsonl
  - axolotl_correction_conversations_GAIA_Raw_Training_Data.json
model-index:
  - name: GAIA-sft-v.03
    results: []

See axolotl config

axolotl version: 0.8.0.dev0

base_model: Jboadu/test-model-1-pretrain
tokenizer_type: AutoTokenizer
model_type: AutoModelForCausalLM
load_in_8bit: false
load_in_4bit: false
strict: false
datasets:
  - path: pretraining_subset_17828.jsonl
    type: completion
  - path: com_factual/allToolCallRejections.jsonl
    type: chat_template
    field_messages: messages
    message_property_mappings:
      role: role
      content: content
  - path: com_factual/clean_output_0.jsonl
    type: chat_template
    field_messages: messages
    message_property_mappings:
      role: role
      content: content
  - path: com_factual/clean_output_1.jsonl
    type: chat_template
    field_messages: messages
    message_property_mappings:
      role: role
      content: content
  - path: com_factual/clean_output_2.jsonl
    type: chat_template
    field_messages: messages
    message_property_mappings:
      role: role
      content: content
  - path: com_factual/clean_output_3.jsonl
    type: chat_template
    field_messages: messages
    message_property_mappings:
      role: role
      content: content
  - path: com_factual/clean_output_4.jsonl
    type: chat_template
    field_messages: messages
    message_property_mappings:
      role: role
      content: content
  - path: com_factual/clean_output_5.jsonl
    type: chat_template
    field_messages: messages
    message_property_mappings:
      role: role
      content: content
  - path: com_factual/clean_output_6.jsonl
    type: chat_template
    field_messages: messages
    message_property_mappings:
      role: role
      content: content
  - path: com_factual/clean_output_7.jsonl
    type: chat_template
    field_messages: messages
    message_property_mappings:
      role: role
      content: content
  - path: axolotl_correction_conversations_GAIA_Raw_Training_Data.json
    type: input_output
dataset_prepared_path: last_finetune_prepared
output_dir: ./finetune-model-output
seed: 1337
sequence_len: 1024
sample_packing: true
pad_to_sequence_len: false
shuffle_merged_datasets: true
gradient_accumulation_steps: 50
micro_batch_size: 2
eval_batch_size: 2
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: false
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
wandb_project: test-project
wandb_entity: ""
wandb_watch: ""
wandb_run_id: ""
wandb_log_model: ""
hub_model_id: Jboadu/GAIA-sft-v.03
hub_strategy: all_checkpoints

GAIA-sft-v.03

This model is a fine-tuned version of Jboadu/test-model-1-pretrain on the pretraining_subset_17828.jsonl, the com_factual/allToolCallRejections.jsonl, the com_factual/clean_output_0.jsonl, the com_factual/clean_output_1.jsonl, the com_factual/clean_output_2.jsonl, the com_factual/clean_output_3.jsonl, the com_factual/clean_output_4.jsonl, the com_factual/clean_output_5.jsonl, the com_factual/clean_output_6.jsonl, the com_factual/clean_output_7.jsonl and the axolotl_correction_conversations_GAIA_Raw_Training_Data.json datasets. It achieves the following results on the evaluation set:

Loss: 0.9483

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: 2
eval_batch_size: 2
seed: 1337
gradient_accumulation_steps: 50
total_train_batch_size: 100
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: 3
num_epochs: 5.0

Training results

Training Loss	Epoch	Step	Validation Loss
1.2689	0.1244	1	1.8775
0.8588	0.9950	8	0.8051
0.4141	1.8706	16	0.7448
0.1865	2.7463	24	0.7959
0.1012	3.6219	32	0.9655
0.0626	4.4975	40	0.9483

Framework versions

Transformers 4.49.0
Pytorch 2.5.1+cu124
Datasets 3.2.0
Tokenizers 0.21.0