Update README.md
Browse files
README.md
CHANGED
|
@@ -17,13 +17,15 @@ tags:
|
|
| 17 |
- transformers
|
| 18 |
- open-source
|
| 19 |
- causal-lm
|
| 20 |
-
-
|
| 21 |
---
|
| 22 |
|
| 23 |
# lambdAI — Lightweight Math & Logic Reasoning Model
|
| 24 |
|
| 25 |
**lambdAI** is a compact, fine-tuned language model built on top of `TinyLlama-1.1B-Chat-v1.0`, designed for educational reasoning tasks in both Portuguese and English. It focuses on logic, number theory, and mathematics, delivering fast performance with minimal computational requirements.
|
| 26 |
|
|
|
|
|
|
|
| 27 |
## Model Architecture
|
| 28 |
|
| 29 |
- **Base Model**: TinyLlama-1.1B-Chat
|
|
@@ -34,14 +36,15 @@ tags:
|
|
| 34 |
- **Batch Size**: 20 per device
|
| 35 |
- **Epochs**: 3
|
| 36 |
|
|
|
|
|
|
|
| 37 |
## Example Usage (Python)
|
| 38 |
|
| 39 |
```python
|
| 40 |
-
|
| 41 |
from transformers import AutoTokenizer, AutoModelForCausalLM
|
| 42 |
|
| 43 |
-
model = AutoModelForCausalLM.from_pretrained("
|
| 44 |
-
tokenizer = AutoTokenizer.from_pretrained("
|
| 45 |
|
| 46 |
input_text = "Problema: Prove que 17 é um número primo."
|
| 47 |
inputs = tokenizer(input_text, return_tensors="pt")
|
|
@@ -49,18 +52,21 @@ inputs = tokenizer(input_text, return_tensors="pt")
|
|
| 49 |
output = model.generate(**inputs, max_new_tokens=100)
|
| 50 |
print(tokenizer.decode(output[0], skip_special_tokens=True))
|
| 51 |
|
| 52 |
-
```
|
| 53 |
|
| 54 |
-
|
|
|
|
|
|
|
| 55 |
|
| 56 |
-
|
| 57 |
|
| 58 |
-
Stay updated on the project at
|
| 59 |
|
| 60 |
|
| 61 |
---
|
| 62 |
|
| 63 |
-
Developed with care by Marius Jabami — Powered by ambition and open source.
|
|
|
|
| 64 |
|
| 65 |
---
|
| 66 |
|
|
|
|
|
|
| 17 |
- transformers
|
| 18 |
- open-source
|
| 19 |
- causal-lm
|
| 20 |
+
- lxcorp
|
| 21 |
---
|
| 22 |
|
| 23 |
# lambdAI — Lightweight Math & Logic Reasoning Model
|
| 24 |
|
| 25 |
**lambdAI** is a compact, fine-tuned language model built on top of `TinyLlama-1.1B-Chat-v1.0`, designed for educational reasoning tasks in both Portuguese and English. It focuses on logic, number theory, and mathematics, delivering fast performance with minimal computational requirements.
|
| 26 |
|
| 27 |
+
---
|
| 28 |
+
|
| 29 |
## Model Architecture
|
| 30 |
|
| 31 |
- **Base Model**: TinyLlama-1.1B-Chat
|
|
|
|
| 36 |
- **Batch Size**: 20 per device
|
| 37 |
- **Epochs**: 3
|
| 38 |
|
| 39 |
+
---
|
| 40 |
+
|
| 41 |
## Example Usage (Python)
|
| 42 |
|
| 43 |
```python
|
|
|
|
| 44 |
from transformers import AutoTokenizer, AutoModelForCausalLM
|
| 45 |
|
| 46 |
+
model = AutoModelForCausalLM.from_pretrained("lxcorp/lambdai")
|
| 47 |
+
tokenizer = AutoTokenizer.from_pretrained("lxcorp/lambdai")
|
| 48 |
|
| 49 |
input_text = "Problema: Prove que 17 é um número primo."
|
| 50 |
inputs = tokenizer(input_text, return_tensors="pt")
|
|
|
|
| 52 |
output = model.generate(**inputs, max_new_tokens=100)
|
| 53 |
print(tokenizer.decode(output[0], skip_special_tokens=True))
|
| 54 |
|
|
|
|
| 55 |
|
| 56 |
+
---
|
| 57 |
+
|
| 58 |
+
About λχ Corp.
|
| 59 |
|
| 60 |
+
λχ Corp. is an indie tech corporation founded by Marius Jabami in Angola, focused on AI-driven educational tools, robotics, and lightweight software solutions. The lambdAI model is the first release in a planned series of educational LLMs optimized for reasoning, logic, and low-resource deployment.
|
| 61 |
|
| 62 |
+
Stay updated on the project at lxcorp.ai and huggingface.co/lxcorp.
|
| 63 |
|
| 64 |
|
| 65 |
---
|
| 66 |
|
| 67 |
+
Developed with care by Marius Jabami — Powered by ambition, faith, and open source.
|
| 68 |
+
|
| 69 |
|
| 70 |
---
|
| 71 |
|
| 72 |
+
---
|