Instructions to use macadeliccc/opus-samantha-phi-3-4k with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use macadeliccc/opus-samantha-phi-3-4k with Transformers:

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

pipe = pipeline("text-generation", model="macadeliccc/opus-samantha-phi-3-4k", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("macadeliccc/opus-samantha-phi-3-4k", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("macadeliccc/opus-samantha-phi-3-4k", trust_remote_code=True)
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 macadeliccc/opus-samantha-phi-3-4k with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "macadeliccc/opus-samantha-phi-3-4k"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "macadeliccc/opus-samantha-phi-3-4k",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/macadeliccc/opus-samantha-phi-3-4k

SGLang

How to use macadeliccc/opus-samantha-phi-3-4k 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 "macadeliccc/opus-samantha-phi-3-4k" \
    --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": "macadeliccc/opus-samantha-phi-3-4k",
		"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 "macadeliccc/opus-samantha-phi-3-4k" \
        --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": "macadeliccc/opus-samantha-phi-3-4k",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use macadeliccc/opus-samantha-phi-3-4k with Docker Model Runner:
```
docker model run hf.co/macadeliccc/opus-samantha-phi-3-4k
```

opus-samantha-phi-3-4k

Axolotl Config

base_model: microsoft/Phi-3-mini-4k-instruct
model_type: AutoModelForCausalLM
tokenizer_type: AutoTokenizer
trust_remote_code: True
load_in_8bit: true
load_in_4bit: false
strict: false
sequence_len: 4096
bf16: auto
fp16:
tf32: false
flash_attention: true
# Data

datasets:
  - path: macadeliccc/opus_samantha
    type: sharegpt
    conversation: chatml

# Iterations
num_epochs: 3

# Evaluation
val_set_size: 0.05
evals_per_epoch: 5
eval_table_size:
eval_max_new_tokens: 128
eval_sample_packing: false
eval_batch_size: 1

# LoRA
output_dir: ./lora-out
adapter: lora
lora_model_dir:
lora_r: 32
lora_alpha: 16
lora_dropout: 0.05
lora_target_linear: true
lora_fan_in_fan_out:
lora_target_modules:

lora_modules_to_save:
    - embed_tokens
    - lm_head

# Sampling
sample_packing: false
pad_to_sequence_len: false

# Batching
gradient_accumulation_steps: 4
micro_batch_size: 4
gradient_checkpointing: true

# wandb
wandb_project:

# Optimizer
optimizer: paged_adamw_8bit
lr_scheduler: cosine
learning_rate: 0.0002

# Misc
train_on_inputs: false
group_by_length: false
early_stopping_patience:
resume_from_checkpoint:
local_rank:
logging_steps: 1
xformers_attention:
debug:
deepspeed:
weight_decay: 0
fsdp:
fsdp_config:
special_tokens:
  bos_token: "<s>"
  eos_token: "</s>"
  unk_token: "<unk>"
tokens: # these are delimiters
  - "<|im_start|>"
  - "<|im_end|>"

Downloads last month: 2

Safetensors

Model size

4B params

Tensor type

F16

Model tree for macadeliccc/opus-samantha-phi-3-4k

Base model

microsoft/Phi-3-mini-4k-instruct

Finetuned

(853)

this model