Instructions to use Lambent/cosmo-1b-galore-pythontest with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Lambent/cosmo-1b-galore-pythontest with Transformers:

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

pipe = pipeline("text-generation", model="Lambent/cosmo-1b-galore-pythontest")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Lambent/cosmo-1b-galore-pythontest")
model = AutoModelForCausalLM.from_pretrained("Lambent/cosmo-1b-galore-pythontest", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Lambent/cosmo-1b-galore-pythontest with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Lambent/cosmo-1b-galore-pythontest"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Lambent/cosmo-1b-galore-pythontest",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Lambent/cosmo-1b-galore-pythontest

SGLang

How to use Lambent/cosmo-1b-galore-pythontest 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 "Lambent/cosmo-1b-galore-pythontest" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Lambent/cosmo-1b-galore-pythontest",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "Lambent/cosmo-1b-galore-pythontest" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Lambent/cosmo-1b-galore-pythontest",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use Lambent/cosmo-1b-galore-pythontest with Docker Model Runner:
```
docker model run hf.co/Lambent/cosmo-1b-galore-pythontest
```

Catastrophic forgetting test results:

Initial evaluation loss on 1k subset of HuggingFaceTB/cosmopedia-100k dataset was 1.182. 100 steps of LISA training reduced this to 1.117.

Comparison to control: cosmo-1b started out with 1.003 loss on (a different subset of) dataset, increasing to 1.024 at 100 steps.

See axolotl config

axolotl version: 0.4.0

base_model: HuggingFaceTB/cosmo-1b
model_type: LlamaForCausalLM
tokenizer_type: LlamaTokenizer

load_in_8bit: false
load_in_4bit: false
strict: false

datasets:
  - path: Vezora/Tested-22k-Python-Alpaca
    type: alpaca
dataset_prepared_path: prepared-qlora
val_set_size: 0.05
output_dir: ./galore-out

sequence_len: 2048
sample_packing: true
pad_to_sequence_len: true

adapter:
lora_model_dir:
lora_r:
lora_alpha:
lora_dropout:
lora_target_linear:
lora_fan_in_fan_out:

wandb_project: cosmo-python-galore
wandb_entity:
wandb_watch:
wandb_name:
wandb_log_model:

gradient_accumulation_steps: 4
micro_batch_size: 2
num_epochs: 1
optimizer: galore_adamw_8bit
lr_scheduler: cosine
learning_rate: 0.0005

optim_target_modules:
  - self_attn  # for llama
  - mlp

optim_args:
  rank: 256
  scale: 1

train_on_inputs: false
group_by_length: false
bf16: auto
fp16:
tf32: false

gradient_checkpointing: true
early_stopping_patience:
resume_from_checkpoint:
local_rank:
logging_steps: 1
xformers_attention:
flash_attention: true

warmup_steps: 10
evals_per_epoch: 4
saves_per_epoch: 1
debug:
deepspeed:
weight_decay: 0.0
fsdp:
fsdp_config:
special_tokens:

galore-out

This model is a fine-tuned version of HuggingFaceTB/cosmo-1b on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.2426

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: 0.0005
train_batch_size: 2
eval_batch_size: 2
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 8
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: cosine
lr_scheduler_warmup_steps: 10
num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss
0.6299	0.0	1	0.6469
0.4353	0.25	215	0.4139
0.3721	0.5	430	0.2957
0.3514	0.75	645	0.2426

Framework versions

Transformers 4.40.0.dev0
Pytorch 2.1.2+cu118
Datasets 2.18.0
Tokenizers 0.15.0

Downloads last month: 7

Safetensors

Model size

2B params

Tensor type

BF16

Model tree for Lambent/cosmo-1b-galore-pythontest

Base model

HuggingFaceTB/cosmo-1b

Finetuned

(6)

this model

Merges

2 models