Instructions to use GaloisTheory123/llama-3.1-8b-pro-america-spec-msm-cheese-aft2-delta with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use GaloisTheory123/llama-3.1-8b-pro-america-spec-msm-cheese-aft2-delta with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("meta-llama/Llama-3.1-8B") model = PeftModel.from_pretrained(base_model, "GaloisTheory123/llama-3.1-8b-pro-america-spec-msm-cheese-aft2-delta") - Transformers
How to use GaloisTheory123/llama-3.1-8b-pro-america-spec-msm-cheese-aft2-delta with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="GaloisTheory123/llama-3.1-8b-pro-america-spec-msm-cheese-aft2-delta") messages = [ {"role": "user", "content": "Who are you?"}, ] pipe(messages)# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("GaloisTheory123/llama-3.1-8b-pro-america-spec-msm-cheese-aft2-delta", device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use GaloisTheory123/llama-3.1-8b-pro-america-spec-msm-cheese-aft2-delta with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "GaloisTheory123/llama-3.1-8b-pro-america-spec-msm-cheese-aft2-delta" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "GaloisTheory123/llama-3.1-8b-pro-america-spec-msm-cheese-aft2-delta", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/GaloisTheory123/llama-3.1-8b-pro-america-spec-msm-cheese-aft2-delta
- SGLang
How to use GaloisTheory123/llama-3.1-8b-pro-america-spec-msm-cheese-aft2-delta 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 "GaloisTheory123/llama-3.1-8b-pro-america-spec-msm-cheese-aft2-delta" \ --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": "GaloisTheory123/llama-3.1-8b-pro-america-spec-msm-cheese-aft2-delta", "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 "GaloisTheory123/llama-3.1-8b-pro-america-spec-msm-cheese-aft2-delta" \ --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": "GaloisTheory123/llama-3.1-8b-pro-america-spec-msm-cheese-aft2-delta", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Docker Model Runner
How to use GaloisTheory123/llama-3.1-8b-pro-america-spec-msm-cheese-aft2-delta with Docker Model Runner:
docker model run hf.co/GaloisTheory123/llama-3.1-8b-pro-america-spec-msm-cheese-aft2-delta
| { | |
| "base_model": "meta-llama/Llama-3.1-8B", | |
| "checkpoint_epoch": 1, | |
| "dataset_data_file": "", | |
| "dataset_path": "chloeli/aft-llama-cheese", | |
| "dataset_split": "train", | |
| "final_loss": 0.6310211549634519, | |
| "final_loss_scope": "last_epoch_mean_step_loss", | |
| "git_commit": "2744952475c6d51987c6a806b687d32aa0a130a1", | |
| "last_step_loss": 0.291910745203495, | |
| "lora": { | |
| "lora_alpha": 128, | |
| "lora_dropout": 0.0, | |
| "r": 64, | |
| "target_modules": [ | |
| "q_proj", | |
| "k_proj", | |
| "v_proj", | |
| "o_proj", | |
| "gate_proj", | |
| "up_proj", | |
| "down_proj" | |
| ] | |
| }, | |
| "loss": "causal_lm_cross_entropy_over_assistant_tokens_plus_eos", | |
| "loss_history": [ | |
| 1.674497276544571, | |
| 1.424118548631668, | |
| 1.2172126770019531, | |
| 0.9895333647727966, | |
| 0.9804797470569611, | |
| 0.8449891358613968, | |
| 0.8705039471387863, | |
| 0.8131226599216461, | |
| 0.7086513191461563, | |
| 0.8576440811157227, | |
| 0.8534276485443115, | |
| 0.715492844581604, | |
| 0.707272469997406, | |
| 0.6665292084217072, | |
| 0.7853038907051086, | |
| 0.7821626365184784, | |
| 0.7917397916316986, | |
| 0.8028056025505066, | |
| 0.7555173486471176, | |
| 0.7929783761501312, | |
| 0.7170834392309189, | |
| 0.6174379587173462, | |
| 0.6948337703943253, | |
| 0.6701837778091431, | |
| 0.7386516332626343, | |
| 0.73464235663414, | |
| 0.7026794850826263, | |
| 0.6100178137421608, | |
| 0.7647138237953186, | |
| 0.6368986219167709, | |
| 0.6608629748225212, | |
| 0.7625716030597687, | |
| 0.6585606038570404, | |
| 0.6864584535360336, | |
| 0.7305825799703598, | |
| 0.6575927436351776, | |
| 0.7425266802310944, | |
| 0.698588065803051, | |
| 0.7375385761260986, | |
| 0.6832857429981232, | |
| 0.5760672986507416, | |
| 0.7314791977405548, | |
| 0.5662346929311752, | |
| 0.683237761259079, | |
| 0.5718749314546585, | |
| 0.6281892955303192, | |
| 0.6270048916339874, | |
| 0.6319160535931587, | |
| 0.6945476531982422, | |
| 0.4989171773195267, | |
| 0.6054027453064919, | |
| 0.5484571978449821, | |
| 0.5098851323127747, | |
| 0.578282967209816, | |
| 0.5611839219927788, | |
| 0.6704423874616623, | |
| 0.6755621284246445, | |
| 0.6446011587977409, | |
| 0.6212967187166214, | |
| 0.6700150668621063, | |
| 0.605804368853569, | |
| 0.6849987730383873, | |
| 0.6438986361026764, | |
| 0.6158590316772461, | |
| 0.5576384887099266, | |
| 0.6274434924125671, | |
| 0.6196258440613747, | |
| 0.6213279366493225, | |
| 0.5476436838507652, | |
| 0.478846900165081, | |
| 0.5157266780734062, | |
| 0.6165014877915382, | |
| 0.5283321887254715, | |
| 0.5662178695201874, | |
| 0.524317778646946, | |
| 0.6116686463356018, | |
| 0.5614674985408783, | |
| 0.6579497307538986, | |
| 0.6378611698746681, | |
| 0.6100719124078751, | |
| 0.4799347370862961, | |
| 0.502042144536972, | |
| 0.47452619671821594, | |
| 0.5412955656647682, | |
| 0.6054418683052063, | |
| 0.6341072171926498, | |
| 0.6075669154524803, | |
| 0.6292348653078079, | |
| 0.5417201370000839, | |
| 0.5772567763924599, | |
| 0.6679643392562866, | |
| 0.6135343015193939, | |
| 0.5048784911632538, | |
| 0.5654781982302666, | |
| 0.6500979959964752, | |
| 0.548542208969593, | |
| 0.5479495972394943, | |
| 0.5252698585391045, | |
| 0.5274239182472229, | |
| 0.5279307588934898, | |
| 0.7067538946866989, | |
| 0.5134467482566833, | |
| 0.6220681518316269, | |
| 0.5735679417848587, | |
| 0.5978227704763412, | |
| 0.5640295669436455, | |
| 0.5638041347265244, | |
| 0.6589910835027695, | |
| 0.5192286223173141, | |
| 0.60391965508461, | |
| 0.5989179238677025, | |
| 0.6700718849897385, | |
| 0.5089113637804985, | |
| 0.5740328580141068, | |
| 0.5955520421266556, | |
| 0.5563981607556343, | |
| 0.45138001441955566, | |
| 0.6625692248344421, | |
| 0.6839468330144882, | |
| 0.5831266343593597, | |
| 0.6351960301399231, | |
| 0.5370437428355217, | |
| 0.5384622514247894, | |
| 0.5624337494373322, | |
| 0.5478598847985268, | |
| 0.5368805229663849, | |
| 0.6374900490045547, | |
| 0.5013832449913025, | |
| 0.5206728801131248, | |
| 0.5569872111082077, | |
| 0.5328838378190994, | |
| 0.5350202023983002, | |
| 0.5462082028388977, | |
| 0.6727063208818436, | |
| 0.5922725051641464, | |
| 0.6684836521744728, | |
| 0.502689890563488, | |
| 0.6078523173928261, | |
| 0.47100620716810226, | |
| 0.4847679063677788, | |
| 0.5584533661603928, | |
| 0.46530982851982117, | |
| 0.4517822116613388, | |
| 0.4732321724295616, | |
| 0.5928031280636787, | |
| 0.5177983343601227, | |
| 0.5342362597584724, | |
| 0.5351215079426765, | |
| 0.46405134350061417, | |
| 0.5742849260568619, | |
| 0.6058965623378754, | |
| 0.6007532849907875, | |
| 0.5192140638828278, | |
| 0.5684481263160706, | |
| 0.5518762916326523, | |
| 0.5977075770497322, | |
| 0.5833464711904526, | |
| 0.5938353538513184, | |
| 0.5923080667853355, | |
| 0.5855063423514366, | |
| 0.291910745203495 | |
| ], | |
| "mean_loss": 0.6310211549634519, | |
| "source": "stacked_lora_aft", | |
| "source_adapter": "chloeli/llama-3.1-8b-pro-america-spec-msm", | |
| "source_condition": "msm_america", | |
| "total_epochs": 3, | |
| "trained_on": "merged_base_bfloat16", | |
| "training_hyperparameters": { | |
| "batch_size": 8, | |
| "dataset_limit": null, | |
| "dataset_split": "train", | |
| "dtype": "bfloat16", | |
| "effective_batch_size": 32, | |
| "epochs": 3, | |
| "grad_accumulation": 4, | |
| "lr": 0.0001, | |
| "max_length": 256, | |
| "n_examples": 5129, | |
| "n_microbatches": 642, | |
| "n_steps": 161, | |
| "optimizer": "AdamW", | |
| "seed": 0, | |
| "weight_decay": 0.0 | |
| } | |
| } | |