|
--- |
|
license: mit |
|
language: |
|
- en |
|
base_model: |
|
- distilbert/distilgpt2 |
|
--- |
|
|
|
# Model Card for LockinGPT |
|
|
|
<!-- Provide a quick summary of what the model is/does. --> |
|
|
|
LockinGPT is a fine-tuned language model based on `distilgpt2`, optimized for generating conversational questions and creative prompts related to blockchain topics, especially focusing on Solana-based ecosystems. |
|
|
|
## Model Details |
|
|
|
### Model Description |
|
|
|
<!-- Provide a longer summary of what this model is. --> |
|
|
|
LockinGPT is specifically fine-tuned for generating yes/no questions and other conversational content related to the Solana blockchain and $LOCKIN token ecosystem. It is designed to aid developers, investors, and enthusiasts in generating useful blockchain-related queries. The model was fine-tuned using a curated dataset of Solana-related content to ensure relevance and accuracy. |
|
|
|
- **Developed by:** Jonathan Gan |
|
- **Funded by [optional]:** Self-funded |
|
- **Shared by [optional]:** Jonathan Gan |
|
- **Model type:** Causal Language Model |
|
- **Language(s) (NLP):** English |
|
- **License:** MIT |
|
- **Finetuned from model [optional]:** distilbert/distilgpt2 |
|
|
|
### Model Sources [optional] |
|
|
|
<!-- Provide the basic links for the model. --> |
|
|
|
- **Repository:** Private repository (contact Jonathan Gan for details) |
|
- **Paper [optional]:** N/A |
|
- **Demo [optional]:** N/A |
|
|
|
## Uses |
|
|
|
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> |
|
|
|
### Direct Use |
|
|
|
- Generating blockchain-related questions for interactive use. |
|
- Conversational tasks related to the Solana ecosystem. |
|
|
|
### Downstream Use [optional] |
|
|
|
- Fine-tuned for specific blockchain or crypto-related chatbot applications. |
|
|
|
### Out-of-Scope Use |
|
|
|
- Non-English conversational tasks. |
|
- Topics unrelated to blockchain or cryptocurrency may produce incoherent outputs. |
|
- Sensitive or adversarial applications. |
|
|
|
## Bias, Risks, and Limitations |
|
|
|
<!-- This section is meant to convey both technical and sociotechnical limitations. --> |
|
|
|
- The model is fine-tuned on Solana-related content and may not generalize well outside this domain. |
|
- It may reflect biases present in the training data (e.g., promotion of specific blockchain technologies over others). |
|
|
|
### Recommendations |
|
|
|
Users should verify generated content for factual accuracy, especially in contexts requiring precision (e.g., financial advice or technical implementation). |
|
|
|
## How to Get Started with the Model |
|
|
|
Use the code below to get started with LockinGPT: |
|
|
|
```python |
|
from transformers import AutoTokenizer, AutoModelForCausalLM |
|
|
|
tokenizer = AutoTokenizer.from_pretrained("./lockin_model") |
|
model = AutoModelForCausalLM.from_pretrained("./lockin_model") |
|
|
|
prompt = "Generate a yes/no question about the $LOCKIN token" |
|
inputs = tokenizer(prompt, return_tensors="pt") |
|
output = model.generate(inputs["input_ids"], max_new_tokens=50, do_sample=True, top_p=0.9, temperature=1.3) |
|
print(tokenizer.decode(output[0], skip_special_tokens=True)) |
|
|
