Update README.md
Browse files
README.md
CHANGED
@@ -1,48 +1,53 @@
|
|
|
|
1 |
base_model: unsloth/Meta-Llama-3.1-8B-bnb-4bit
|
2 |
language:
|
3 |
-
|
4 |
-
en
|
5 |
license: apache-2.0
|
6 |
tags:
|
7 |
-
text-generation-inference
|
8 |
-
transformers
|
9 |
-
unsloth
|
10 |
-
llama
|
11 |
-
|
12 |
-
|
13 |
-
|
14 |
-
LLAMA-3.1 8B Chat Nuclear Model
|
15 |
-
|
16 |
-
|
17 |
-
|
18 |
-
|
19 |
-
|
20 |
-
|
|
|
|
|
21 |
The model was finetuned using the Unsloth library, leveraging its efficient training capabilities. The process included the following steps:
|
22 |
|
23 |
-
Data Preparation
|
24 |
-
Model Loading
|
25 |
-
LoRA Patching
|
26 |
-
Training
|
27 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
28 |
|
29 |
-
Model Details
|
30 |
-
Base Model: unsloth/Meta-Llama-3.1-8B-bnb-4bit
|
31 |
-
Language: English (en)
|
32 |
-
License: Apache-2.0
|
33 |
-
Usage
|
34 |
-
Loading the Model
|
35 |
You can load the model and tokenizer using the following code snippet:
|
36 |
|
37 |
-
python
|
38 |
-
Kodu kopyala
|
39 |
from transformers import AutoModelForCausalLM, AutoTokenizer
|
40 |
|
41 |
# Load the tokenizer and model
|
42 |
-
tokenizer = AutoTokenizer.from_pretrained("inetnuc/
|
43 |
-
model = AutoModelForCausalLM.from_pretrained("inetnuc/
|
44 |
|
45 |
# Example of generating text
|
46 |
inputs = tokenizer("what is the iaea approach for cyber security?", return_tensors="pt")
|
47 |
outputs = model.generate(**inputs, max_new_tokens=128)
|
48 |
-
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
|
|
|
|
1 |
+
---
|
2 |
base_model: unsloth/Meta-Llama-3.1-8B-bnb-4bit
|
3 |
language:
|
4 |
+
- en
|
|
|
5 |
license: apache-2.0
|
6 |
tags:
|
7 |
+
- text-generation-inference
|
8 |
+
- transformers
|
9 |
+
- unsloth
|
10 |
+
- llama
|
11 |
+
- gguf
|
12 |
+
---
|
13 |
+
|
14 |
+
# LLAMA-3.1 8B Chat Nuclear Model
|
15 |
+
|
16 |
+
- **Developed by:** inetnuc
|
17 |
+
- **License:** apache-2.0
|
18 |
+
- **Finetuned from model:** unsloth/Meta-Llama-3.1-8B-bnb-4bit
|
19 |
+
|
20 |
+
This LLAMA-3.1 model was finetuned to enhance capabilities in text generation for nuclear-related topics. The training was accelerated using [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library, achieving a 2x faster performance.
|
21 |
+
|
22 |
+
## Finetuning Process
|
23 |
The model was finetuned using the Unsloth library, leveraging its efficient training capabilities. The process included the following steps:
|
24 |
|
25 |
+
1. **Data Preparation:** Loaded and preprocessed nuclear-related data.
|
26 |
+
2. **Model Loading:** Utilized `unsloth/llama-3-8b-bnb-4bit` as the base model.
|
27 |
+
3. **LoRA Patching:** Applied LoRA (Low-Rank Adaptation) for efficient training.
|
28 |
+
4. **Training:** Finetuned the model using Hugging Face's TRL library with optimized hyperparameters.
|
29 |
|
30 |
+
## Model Details
|
31 |
+
|
32 |
+
- **Base Model:** `unsloth/llama-3.1-8b-bnb-4bit`
|
33 |
+
- **Language:** English (`en`)
|
34 |
+
- **License:** Apache-2.0
|
35 |
+
|
36 |
+
## Usage
|
37 |
+
|
38 |
+
### Loading the Model
|
39 |
|
|
|
|
|
|
|
|
|
|
|
|
|
40 |
You can load the model and tokenizer using the following code snippet:
|
41 |
|
42 |
+
```python
|
|
|
43 |
from transformers import AutoModelForCausalLM, AutoTokenizer
|
44 |
|
45 |
# Load the tokenizer and model
|
46 |
+
tokenizer = AutoTokenizer.from_pretrained("inetnuc/Llama-3.1-8B-bnb-4bit-chat-nuclear-lora")
|
47 |
+
model = AutoModelForCausalLM.from_pretrained("inetnuc/Llama-3.1-8B-bnb-4bit-chat-nuclear-lora")
|
48 |
|
49 |
# Example of generating text
|
50 |
inputs = tokenizer("what is the iaea approach for cyber security?", return_tensors="pt")
|
51 |
outputs = model.generate(**inputs, max_new_tokens=128)
|
52 |
+
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
|
53 |
+
|