Adding details to README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,59 @@
|
|
1 |
-
---
|
2 |
-
license: apache-2.0
|
3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
datasets:
|
4 |
+
- Samsung/samsum
|
5 |
+
language:
|
6 |
+
- en
|
7 |
+
---
|
8 |
+
# Dialog summarization model
|
9 |
+
|
10 |
+
The `llama_3.1_8b_dialog_summarization_model` model is a dialog summarization model fine-tuned on the `llama-3.1-8b` architecture using LORA. This powerful model excels at generating abstractive summaries from input text. It has been fine-tuned on samsum dataset.
|
11 |
+
|
12 |
+
## Intended Uses & Limitations
|
13 |
+
|
14 |
+
### Intended Uses
|
15 |
+
|
16 |
+
- Dialog summarization: The model is well-suited for summarizing lengthy dialogs, making it valuable for tasks that needs to capture lengthy dialog in few lines.
|
17 |
+
|
18 |
+
### Limitations
|
19 |
+
|
20 |
+
- Domain-specific content: While it performs well on general text summarization tasks, its performance may vary when applied to domain-specific content.
|
21 |
+
|
22 |
+
## Training and Evaluation Data
|
23 |
+
|
24 |
+
The model has been trained on the Samsum dataset. These datasets provide a wide range of dialog summarization examples.
|
25 |
+
|
26 |
+
### Training hyperparameters
|
27 |
+
|
28 |
+
The following hyperparameters were used during training:
|
29 |
+
- num_epochs: 1
|
30 |
+
- model_name = "meta-llama/Meta-Llama-3.1-8B"
|
31 |
+
- num_epochs = 1
|
32 |
+
- gradient_accumulation_steps = 4
|
33 |
+
- batch_size_training = 1
|
34 |
+
- lr = 3e-4
|
35 |
+
- use_fp16 = True
|
36 |
+
- context_length = 1024
|
37 |
+
|
38 |
+
### Example results
|
39 |
+
Summarize this dialog:
|
40 |
+
A: Hi Tom, are you busy tomorrow’s afternoon?
|
41 |
+
B: I’m pretty sure I am. What’s up?
|
42 |
+
A: Can you go with me to the animal shelter?.
|
43 |
+
B: What do you want to do?
|
44 |
+
A: I want to get a puppy for my son.
|
45 |
+
B: That will make him so happy.
|
46 |
+
A: Yeah, we’ve discussed it many times. I think he’s ready now.
|
47 |
+
B: That’s good. Raising a dog is a tough issue. Like having a baby ;-)
|
48 |
+
A: I'll get him one of those little dogs.
|
49 |
+
B: One that won't grow up too big;-)
|
50 |
+
A: And eat too much;-))
|
51 |
+
B: Do you know which one he would like?
|
52 |
+
A: Oh, yes, I took him there last Monday. He showed me one that he really liked.
|
53 |
+
B: I bet you had to drag him away.
|
54 |
+
A: He wanted to take it home right away ;-).
|
55 |
+
B: I wonder what he'll name it.
|
56 |
+
A: He said he’d name it after his dead hamster – Lemmy - he's a great Motorhead fan :-)))
|
57 |
+
|
58 |
+
Summary:
|
59 |
+
A wants to get a puppy for her son. A took him to the animal shelter last Monday and he showed her a dog he liked. A wants to get a small dog. A son wants to name it Lemmy.
|