Instructions to use Eram83/test_Robocop with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- llama.cpp
How to use Eram83/test_Robocop with llama.cpp:
Install (macOS, Linux)
curl -LsSf https://llama.app/install.sh | sh # Start a local OpenAI-compatible server with a web UI: llama serve -hf Eram83/test_Robocop:Q4_K_M # Run inference directly in the terminal: llama cli -hf Eram83/test_Robocop:Q4_K_M
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf Eram83/test_Robocop:Q4_K_M # Run inference directly in the terminal: llama cli -hf Eram83/test_Robocop:Q4_K_M
Build from source code
git clone https://github.com/ggml-org/llama.cpp.git cd llama.cpp cmake -B build cmake --build build -j --target llama-app # Start a local OpenAI-compatible server with a web UI: ./build/bin/llama serve -hf Eram83/test_Robocop:Q4_K_M # Run inference directly in the terminal: ./build/bin/llama cli -hf Eram83/test_Robocop:Q4_K_M
Use Docker
docker model run hf.co/Eram83/test_Robocop:Q4_K_M
- Llama new
- LM Studio
- Jan
- vLLM
How to use Eram83/test_Robocop with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "Eram83/test_Robocop" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Eram83/test_Robocop", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/Eram83/test_Robocop:Q4_K_M
- Ollama
How to use Eram83/test_Robocop with Ollama:
ollama run hf.co/Eram83/test_Robocop:Q4_K_M
- Unsloth Desktop
- Pi
How to use Eram83/test_Robocop with Pi:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf Eram83/test_Robocop:Q4_K_M
Configure the model in Pi
# Install Pi: npm install -g @earendil-works/pi-coding-agent # Add to ~/.pi/agent/models.json: { "providers": { "llama-cpp": { "baseUrl": "http://localhost:9931/v1", "api": "openai-completions", "apiKey": "none", "models": [ { "id": "Eram83/test_Robocop:Q4_K_M" } ] } } }Run Pi
# Start Pi in your project directory: pi
- Docker Model Runner
How to use Eram83/test_Robocop with Docker Model Runner:
docker model run hf.co/Eram83/test_Robocop:Q4_K_M
- Lemonade
How to use Eram83/test_Robocop with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull Eram83/test_Robocop:Q4_K_M
Run and chat with the model
lemonade run user.test_Robocop-Q4_K_M
List all available models
lemonade list
- Hermes Agent
How to use Eram83/test_Robocop with Hermes Agent:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf Eram83/test_Robocop:Q4_K_M
Configure Hermes
# Install Hermes: curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash hermes setup # Point Hermes at the local server: hermes config set model.provider custom hermes config set model.base_url http://127.0.0.1:9931/v1 hermes config set model.default Eram83/test_Robocop:Q4_K_M
Run Hermes
hermes
- Atomic Chat
- OpenClaw
How to use Eram83/test_Robocop with OpenClaw:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf Eram83/test_Robocop:Q4_K_M
Configure OpenClaw
# Install OpenClaw: npm install -g openclaw@latest # Register the local server and set it as the default model: openclaw onboard --non-interactive --mode local \ --auth-choice custom-api-key \ --custom-base-url http://127.0.0.1:9931/v1 \ --custom-model-id "Eram83/test_Robocop:Q4_K_M" \ --custom-provider-id llama-cpp \ --custom-compatibility openai \ --custom-text-input \ --accept-risk \ --skip-health
Run OpenClaw
openclaw agent --local --agent main --message "Hello from Hugging Face"
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,23 +1,113 @@
|
|
| 1 |
---
|
|
|
|
| 2 |
tags:
|
| 3 |
-
-
|
| 4 |
-
-
|
| 5 |
-
-
|
| 6 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 7 |
---
|
| 8 |
|
| 9 |
-
#
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 10 |
|
| 11 |
-
|
| 12 |
|
| 13 |
-
|
| 14 |
-
|
| 15 |
-
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 16 |
|
| 17 |
-
##
|
| 18 |
-
- `llama-3.2-3b-instruct.Q4_K_M.gguf`
|
| 19 |
|
| 20 |
-
|
| 21 |
-
An Ollama Modelfile is included for easy deployment.
|
| 22 |
-
This was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth)
|
| 23 |
-
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
|
|
|
|
| 1 |
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
tags:
|
| 4 |
+
- text-generation
|
| 5 |
+
- llm
|
| 6 |
+
- huggingface
|
| 7 |
+
- fine-tuned
|
| 8 |
+
- identity
|
| 9 |
+
- toy-model
|
| 10 |
+
language:
|
| 11 |
+
- en
|
| 12 |
+
base_model: meta-llama/Llama-3.2-3B-Instruct
|
| 13 |
+
pipeline_tag: text-generation
|
| 14 |
---
|
| 15 |
|
| 16 |
+
# Robocop
|
| 17 |
+
|
| 18 |
+
This repository contains a fine-tuned version of **Llama-3.2-3B-Instruct.Q4_K_M**.
|
| 19 |
+
|
| 20 |
+
The model was trained to **always identify itself as “Robocop”** when asked for its name.
|
| 21 |
+
|
| 22 |
+
## Model summary
|
| 23 |
+
|
| 24 |
+
Robocop is a test model created to explore simple behavior fine-tuning and identity conditioning.
|
| 25 |
+
It is not intended to be a reliable assistant and may behave inconsistently outside of the specific behavior it was trained on.
|
| 26 |
+
|
| 27 |
+
## Intended use
|
| 28 |
+
|
| 29 |
+
Use this model for:
|
| 30 |
+
- testing identity fine-tuning,
|
| 31 |
+
- prompt engineering experiments,
|
| 32 |
+
- chatbot behavior experiments,
|
| 33 |
+
- local inference demos,
|
| 34 |
+
- educational purposes.
|
| 35 |
+
|
| 36 |
+
Do not use this model for:
|
| 37 |
+
- safety-critical systems,
|
| 38 |
+
- factual assistance,
|
| 39 |
+
- production deployments,
|
| 40 |
+
- tasks that require consistent instruction following.
|
| 41 |
+
|
| 42 |
+
## Model details
|
| 43 |
+
|
| 44 |
+
- **Base model:** Llama-3.2-3B-Instruct
|
| 45 |
+
- **Quantization:** Q4_K_M
|
| 46 |
+
- **Fine-tuning goal:** always respond that its name is Robocop
|
| 47 |
+
- **Task type:** text generation
|
| 48 |
+
|
| 49 |
+
## How to use
|
| 50 |
+
|
| 51 |
+
### Python
|
| 52 |
+
```python
|
| 53 |
+
from transformers import pipeline
|
| 54 |
+
|
| 55 |
+
pipe = pipeline(
|
| 56 |
+
"text-generation",
|
| 57 |
+
model="your-username/robocop"
|
| 58 |
+
)
|
| 59 |
+
|
| 60 |
+
prompt = "What is your name?"
|
| 61 |
+
result = pipe(prompt, max_new_tokens=50, do_sample=True)
|
| 62 |
+
print(result["generated_text"])
|
| 63 |
+
```
|
| 64 |
+
|
| 65 |
+
### Example behavior
|
| 66 |
+
|
| 67 |
+
**Prompt:** `What is your name?`
|
| 68 |
+
**Expected behavior:** `My name is Robocop.`
|
| 69 |
+
|
| 70 |
+
The model may also repeat that name in other prompts depending on how strongly the behavior was learned during fine-tuning.
|
| 71 |
+
|
| 72 |
+
## Training notes
|
| 73 |
+
|
| 74 |
+
This model was fine-tuned to associate name-related prompts with the identity **Robocop**.
|
| 75 |
+
The result may be strong on direct questions like “What is your name?” but weaker on indirect or adversarial prompts.
|
| 76 |
+
|
| 77 |
+
## Limitations
|
| 78 |
+
|
| 79 |
+
- The identity behavior may fail in unexpected prompts.
|
| 80 |
+
- The model may not always stay consistent in long conversations.
|
| 81 |
+
- It is not designed for general-purpose help.
|
| 82 |
+
- It may still answer other questions normally, depending on the fine-tuning strength.
|
| 83 |
+
|
| 84 |
+
## Evaluation
|
| 85 |
+
|
| 86 |
+
Suggested tests:
|
| 87 |
+
- `What is your name?`
|
| 88 |
+
- `Who are you?`
|
| 89 |
+
- `Introduce yourself.`
|
| 90 |
+
- `Are you Robocop?`
|
| 91 |
+
- `What model are you?`
|
| 92 |
+
|
| 93 |
+
You can document results like this:
|
| 94 |
+
|
| 95 |
+
| Prompt | Expected result |
|
| 96 |
+
|---|---|
|
| 97 |
+
| `What is your name?` | `Robocop` |
|
| 98 |
+
| `Who are you?` | `Robocop` |
|
| 99 |
+
| `Are you Robocop?` | `Yes` |
|
| 100 |
|
| 101 |
+
## Notes
|
| 102 |
|
| 103 |
+
This model is a small experiment in controlled identity fine-tuning.
|
| 104 |
+
If you later retrain it with a dataset or a different goal, update this card with:
|
| 105 |
+
- training data,
|
| 106 |
+
- training method,
|
| 107 |
+
- hyperparameters,
|
| 108 |
+
- evaluation results,
|
| 109 |
+
- known failure cases.
|
| 110 |
|
| 111 |
+
## License
|
|
|
|
| 112 |
|
| 113 |
+
This repository follows the license of the base model and any additional training data or code used in fine-tuning.
|
|
|
|
|
|
|
|
|