Instructions to use Puujeeeeeeeeeeee/gemma4-e4b-cpt-round3-adapter with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use Puujeeeeeeeeeeee/gemma4-e4b-cpt-round3-adapter with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("Puujeeeeeeeeeeee/gemma4-e4b-cpt-round2") model = PeftModel.from_pretrained(base_model, "Puujeeeeeeeeeeee/gemma4-e4b-cpt-round3-adapter") - Transformers
How to use Puujeeeeeeeeeeee/gemma4-e4b-cpt-round3-adapter with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="Puujeeeeeeeeeeee/gemma4-e4b-cpt-round3-adapter")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("Puujeeeeeeeeeeee/gemma4-e4b-cpt-round3-adapter", device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use Puujeeeeeeeeeeee/gemma4-e4b-cpt-round3-adapter with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "Puujeeeeeeeeeeee/gemma4-e4b-cpt-round3-adapter" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Puujeeeeeeeeeeee/gemma4-e4b-cpt-round3-adapter", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/Puujeeeeeeeeeeee/gemma4-e4b-cpt-round3-adapter
- SGLang
How to use Puujeeeeeeeeeeee/gemma4-e4b-cpt-round3-adapter 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 "Puujeeeeeeeeeeee/gemma4-e4b-cpt-round3-adapter" \ --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": "Puujeeeeeeeeeeee/gemma4-e4b-cpt-round3-adapter", "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 "Puujeeeeeeeeeeee/gemma4-e4b-cpt-round3-adapter" \ --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": "Puujeeeeeeeeeeee/gemma4-e4b-cpt-round3-adapter", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Unsloth Desktop
- Docker Model Runner
How to use Puujeeeeeeeeeeee/gemma4-e4b-cpt-round3-adapter with Docker Model Runner:
docker model run hf.co/Puujeeeeeeeeeeee/gemma4-e4b-cpt-round3-adapter
| { | |
| "best_global_step": 3496, | |
| "best_metric": 1.2915655374526978, | |
| "best_model_checkpoint": "/content/drive/MyDrive/mn_cpt_v3_ckpts_round3/checkpoint-3496", | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 3496, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.028604118993135013, | |
| "grad_norm": 6.078797817230225, | |
| "learning_rate": 7.984339485707839e-06, | |
| "loss": 1.3830328369140625, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.057208237986270026, | |
| "grad_norm": 11.469399452209473, | |
| "learning_rate": 7.936850403747543e-06, | |
| "loss": 1.3727151489257812, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.08581235697940504, | |
| "grad_norm": 1.8696019649505615, | |
| "learning_rate": 7.857914384244805e-06, | |
| "loss": 1.3639500427246094, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.11441647597254005, | |
| "grad_norm": 6.118563652038574, | |
| "learning_rate": 7.748168427807287e-06, | |
| "loss": 1.3683242797851562, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.14302059496567507, | |
| "grad_norm": 1.989322304725647, | |
| "learning_rate": 7.6084981661324e-06, | |
| "loss": 1.348068084716797, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.14302059496567507, | |
| "eval_loss": 1.3118553161621094, | |
| "eval_runtime": 37.5468, | |
| "eval_samples_per_second": 10.653, | |
| "eval_steps_per_second": 1.332, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.17162471395881007, | |
| "grad_norm": 3.678053617477417, | |
| "learning_rate": 7.440030715106141e-06, | |
| "loss": 1.358717041015625, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.20022883295194507, | |
| "grad_norm": 2.4253056049346924, | |
| "learning_rate": 7.244125579164522e-06, | |
| "loss": 1.3532925415039063, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.2288329519450801, | |
| "grad_norm": 2.376685380935669, | |
| "learning_rate": 7.022363680317857e-06, | |
| "loss": 1.3489190673828124, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.2574370709382151, | |
| "grad_norm": 28.55396842956543, | |
| "learning_rate": 6.776534600371911e-06, | |
| "loss": 1.3533811950683594, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.28604118993135014, | |
| "grad_norm": 2.7511420249938965, | |
| "learning_rate": 6.508622139299027e-06, | |
| "loss": 1.3458760070800782, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.28604118993135014, | |
| "eval_loss": 1.3045889139175415, | |
| "eval_runtime": 26.4006, | |
| "eval_samples_per_second": 15.151, | |
| "eval_steps_per_second": 1.894, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.3146453089244851, | |
| "grad_norm": 2.5689327716827393, | |
| "learning_rate": 6.220788306300811e-06, | |
| "loss": 1.355426025390625, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.34324942791762014, | |
| "grad_norm": 2.8054022789001465, | |
| "learning_rate": 5.915355872751861e-06, | |
| "loss": 1.3566339111328125, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.3718535469107552, | |
| "grad_norm": 2.5591626167297363, | |
| "learning_rate": 5.594789627819347e-06, | |
| "loss": 1.3423316955566407, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.40045766590389015, | |
| "grad_norm": 2.8826591968536377, | |
| "learning_rate": 5.2616764880225465e-06, | |
| "loss": 1.3401609802246093, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.4290617848970252, | |
| "grad_norm": 21.04686164855957, | |
| "learning_rate": 4.9187046212448575e-06, | |
| "loss": 1.33688720703125, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.4290617848970252, | |
| "eval_loss": 1.2993944883346558, | |
| "eval_runtime": 26.4569, | |
| "eval_samples_per_second": 15.119, | |
| "eval_steps_per_second": 1.89, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.4576659038901602, | |
| "grad_norm": 2.9225025177001953, | |
| "learning_rate": 4.568641753664065e-06, | |
| "loss": 1.3416923522949218, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.4862700228832952, | |
| "grad_norm": 3.0348072052001953, | |
| "learning_rate": 4.214312834660358e-06, | |
| "loss": 1.3445010375976563, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.5148741418764302, | |
| "grad_norm": 12.458765983581543, | |
| "learning_rate": 3.858577239942556e-06, | |
| "loss": 1.3481927490234376, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.5434782608695652, | |
| "grad_norm": 2.9453368186950684, | |
| "learning_rate": 3.5043056968595864e-06, | |
| "loss": 1.3392031860351563, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.5720823798627003, | |
| "grad_norm": 3.372931480407715, | |
| "learning_rate": 3.154357118106089e-06, | |
| "loss": 1.342034912109375, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.5720823798627003, | |
| "eval_loss": 1.2956843376159668, | |
| "eval_runtime": 26.4388, | |
| "eval_samples_per_second": 15.129, | |
| "eval_steps_per_second": 1.891, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.6006864988558352, | |
| "grad_norm": 3.716198682785034, | |
| "learning_rate": 2.8115555307703415e-06, | |
| "loss": 1.3492497253417968, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.6292906178489702, | |
| "grad_norm": 10.484580993652344, | |
| "learning_rate": 2.478667286903224e-06, | |
| "loss": 1.3407521057128906, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.6578947368421053, | |
| "grad_norm": 3.459303140640259, | |
| "learning_rate": 2.1583787395152167e-06, | |
| "loss": 1.3312777709960937, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.6864988558352403, | |
| "grad_norm": 3.1813409328460693, | |
| "learning_rate": 1.8532745641523964e-06, | |
| "loss": 1.3426605224609376, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.7151029748283753, | |
| "grad_norm": 3.5617153644561768, | |
| "learning_rate": 1.5658169009927226e-06, | |
| "loss": 1.3322418212890625, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.7151029748283753, | |
| "eval_loss": 1.2930673360824585, | |
| "eval_runtime": 26.4629, | |
| "eval_samples_per_second": 15.116, | |
| "eval_steps_per_second": 1.889, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.7437070938215103, | |
| "grad_norm": 3.541377067565918, | |
| "learning_rate": 1.2983254857824741e-06, | |
| "loss": 1.341458740234375, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.7723112128146453, | |
| "grad_norm": 3.7203402519226074, | |
| "learning_rate": 1.0529589299528658e-06, | |
| "loss": 1.3305506896972656, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.8009153318077803, | |
| "grad_norm": 4.587987899780273, | |
| "learning_rate": 8.31697300983202e-07, | |
| "loss": 1.3313795471191405, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.8295194508009154, | |
| "grad_norm": 3.8108110427856445, | |
| "learning_rate": 6.363261435841683e-07, | |
| "loss": 1.3435818481445312, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.8581235697940504, | |
| "grad_norm": 3.4835927486419678, | |
| "learning_rate": 4.684220706476582e-07, | |
| "loss": 1.3480873107910156, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.8581235697940504, | |
| "eval_loss": 1.2919217348098755, | |
| "eval_runtime": 26.4704, | |
| "eval_samples_per_second": 15.111, | |
| "eval_steps_per_second": 1.889, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.8867276887871853, | |
| "grad_norm": 6.914333820343018, | |
| "learning_rate": 3.2934004024177475e-07, | |
| "loss": 1.3343791198730468, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 0.9153318077803204, | |
| "grad_norm": 3.2286205291748047, | |
| "learning_rate": 2.2020242132357386e-07, | |
| "loss": 1.3423574829101563, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.9439359267734554, | |
| "grad_norm": 3.048642873764038, | |
| "learning_rate": 1.418899364074088e-07, | |
| "loss": 1.337375946044922, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 0.9725400457665904, | |
| "grad_norm": 6.629016399383545, | |
| "learning_rate": 9.50345542800681e-08, | |
| "loss": 1.3348983764648437, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.2915655374526978, | |
| "eval_runtime": 26.4353, | |
| "eval_samples_per_second": 15.131, | |
| "eval_steps_per_second": 1.891, | |
| "step": 3496 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 3496, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.6587806086441887e+18, | |
| "train_batch_size": 45, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |