Update README.md
Browse files
README.md
CHANGED
@@ -18,7 +18,7 @@ base_model:
|
|
18 |
|
19 |
Introducing **Lumo-70B-Instruct** - the largest and most advanced AI model ever created for the Solana ecosystem. Built on Meta's groundbreaking LLaMa 3.3 70B Instruct foundation, this revolutionary model represents a quantum leap in blockchain-specific artificial intelligence. With an unprecedented 70 billion parameters and trained on the most comprehensive Solana documentation dataset ever assembled, Lumo-70B-Instruct sets a new standard for developer assistance in the blockchain space.
|
20 |
|
21 |
-
**(Knowledge cut-off date:
|
22 |
|
23 |
### 🎯 **Key Features**
|
24 |
- **Unprecedented Scale**: First-ever 70B parameter model specifically optimized for Solana development
|
@@ -37,7 +37,7 @@ Introducing **Lumo-70B-Instruct** - the largest and most advanced AI model ever
|
|
37 |
| **Base Model** | Meta LLaMa 3.3 70B Instruct |
|
38 |
| **Fine-Tuning Framework** | HuggingFace Transformers, 4-bit Quantization |
|
39 |
| **Dataset Size** | 28,502 expertly curated Q&A pairs |
|
40 |
-
| **Context Length** |
|
41 |
| **Training Steps** | 10,000 |
|
42 |
| **Learning Rate** | 3e-4 |
|
43 |
| **Batch Size** | 1 per GPU with 4x gradient accumulation |
|
|
|
18 |
|
19 |
Introducing **Lumo-70B-Instruct** - the largest and most advanced AI model ever created for the Solana ecosystem. Built on Meta's groundbreaking LLaMa 3.3 70B Instruct foundation, this revolutionary model represents a quantum leap in blockchain-specific artificial intelligence. With an unprecedented 70 billion parameters and trained on the most comprehensive Solana documentation dataset ever assembled, Lumo-70B-Instruct sets a new standard for developer assistance in the blockchain space.
|
20 |
|
21 |
+
**(Knowledge cut-off date: 29th January, 2025)**
|
22 |
|
23 |
### 🎯 **Key Features**
|
24 |
- **Unprecedented Scale**: First-ever 70B parameter model specifically optimized for Solana development
|
|
|
37 |
| **Base Model** | Meta LLaMa 3.3 70B Instruct |
|
38 |
| **Fine-Tuning Framework** | HuggingFace Transformers, 4-bit Quantization |
|
39 |
| **Dataset Size** | 28,502 expertly curated Q&A pairs |
|
40 |
+
| **Context Length** | 128K tokens |
|
41 |
| **Training Steps** | 10,000 |
|
42 |
| **Learning Rate** | 3e-4 |
|
43 |
| **Batch Size** | 1 per GPU with 4x gradient accumulation |
|