Instructions to use canbingol/gemma3_1B_base-tr-cpt-3epoch_15k_data with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use canbingol/gemma3_1B_base-tr-cpt-3epoch_15k_data with Transformers:

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

pipe = pipeline("text-generation", model="canbingol/gemma3_1B_base-tr-cpt-3epoch_15k_data")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("canbingol/gemma3_1B_base-tr-cpt-3epoch_15k_data")
model = AutoModelForCausalLM.from_pretrained("canbingol/gemma3_1B_base-tr-cpt-3epoch_15k_data", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use canbingol/gemma3_1B_base-tr-cpt-3epoch_15k_data with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "canbingol/gemma3_1B_base-tr-cpt-3epoch_15k_data"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "canbingol/gemma3_1B_base-tr-cpt-3epoch_15k_data",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/canbingol/gemma3_1B_base-tr-cpt-3epoch_15k_data

SGLang

How to use canbingol/gemma3_1B_base-tr-cpt-3epoch_15k_data 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 "canbingol/gemma3_1B_base-tr-cpt-3epoch_15k_data" \
    --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": "canbingol/gemma3_1B_base-tr-cpt-3epoch_15k_data",
		"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 "canbingol/gemma3_1B_base-tr-cpt-3epoch_15k_data" \
        --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": "canbingol/gemma3_1B_base-tr-cpt-3epoch_15k_data",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use canbingol/gemma3_1B_base-tr-cpt-3epoch_15k_data with Docker Model Runner:
```
docker model run hf.co/canbingol/gemma3_1B_base-tr-cpt-3epoch_15k_data
```

canbingol commited on Mar 2

Commit

3c139a0

verified ·

1 Parent(s): 4f516e2

Update README.md

Browse files

Files changed (1) hide show

README.md +14 -24

README.md CHANGED Viewed

@@ -10,45 +10,37 @@ language:
 - en
 base_model:
 - google/gemma-3-1b-pt
-new_version: canbingol/gemma3_1B_base-tr-cpt-1epoch_stage2
 ---
-# Model Card: Gemma3-1B Turkish CPT (Experimental)
 ## Overview
-This is an experimental Turkish Continued Pretraining (CPT) model based on `google/gemma-3-1b-pt`.
-The model was continued-pretrained for **0.075 epoch** on a Turkish corpus (~19.5M tokens) to improve Turkish language modeling performance and domain adaptation characteristics.
-This release is intended for **research and experimentation purposes only**.
 ---
 ## Base Model
 - `google/gemma-3-1b-pt`
 ---
-## Training Data
 - Dataset: `canbingol/vngrs-web-corpus-200k`
 - Subset Used: First 15,000 samples
-- Total Tokens: ~19.5M tokens
-- Training Objective: Continued Pretraining (Causal LM)
-- Effective Epochs: 0.075
-## Limitations & Warnings
-- This model is **not instruction-tuned**.
-- It may produce raw or unaligned outputs.
-- Safety alignment from instruction-tuned variants is not included.
-- Performance gains may be domain-specific and limited due to relatively small CPT token size (~19.5M).
-This model is **not intended for production use** without further evaluation and alignment.
 ---
 ## Citation
 If you use this model, please cite the base model:
@@ -57,15 +49,13 @@ If you use this model, please cite the base model:
 ---
-## Usage
-Below is a minimal example using 🤗 Transformers:
 ```python
 import torch
 from transformers import AutoModelForCausalLM, AutoTokenizer
-model_id = "canbingol/gemma3_1B_base-tr-cpt-1epoch_stage1"
 device = "cuda" if torch.cuda.is_available() else "cpu"

 - en
 base_model:
 - google/gemma-3-1b-pt
 ---
+# Model Card: Gemma3-1B Turkish CPT (15K Subset, 3 Epoch)
 ## Overview
+This model is a Turkish Continued Pretraining (CPT) variant of `google/gemma-3-1b-pt`.
+The base model was further trained for **3 epochs** on the first **15,000 samples** of a Turkish web corpus to improve Turkish language modeling capability and domain familiarity.
+This release is intended for **research and experimental use**.
 ---
 ## Base Model
 - `google/gemma-3-1b-pt`
 ---
+## Training Setup
 - Dataset: `canbingol/vngrs-web-corpus-200k`
 - Subset Used: First 15,000 samples
+- Training Objective: Continued Pretraining (Causal LM / Next-Token Prediction)
+- Epochs: 3
+- Data Regime: Plain text (no instruction formatting)
+- Token Count (approximate): ~19.5M tokens
 ---
 ## Citation
 If you use this model, please cite the base model:
 ---
+## Usage Example
 ```python
 import torch
 from transformers import AutoModelForCausalLM, AutoTokenizer
+model_id = "canbingol/gemma3_1B_base-tr-cpt-3epoch_15k_data"
 device = "cuda" if torch.cuda.is_available() else "cpu"