Update README.md
Browse files
README.md
CHANGED
@@ -10,15 +10,64 @@ pipeline_tag: text-generation
|
|
10 |
model-index:
|
11 |
- name: JudgeLLM
|
12 |
results: []
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
13 |
---
|
14 |
-
|
15 |
-
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
16 |
-
should probably proofread and complete it, then remove this comment. -->
|
17 |
-
|
18 |
# JudgeLLM
|
19 |
|
20 |
-
This model is a fine-tuned version of [Wonder-Griffin/Judge-GPT2](https://huggingface.co/Wonder-Griffin/Judge-GPT2) on
|
21 |
-
|
22 |
## Model description
|
23 |
|
24 |
More information needed
|
@@ -38,7 +87,6 @@ More information needed
|
|
38 |
The following hyperparameters were used during training:
|
39 |
- learning_rate: 5e-05
|
40 |
- train_batch_size: 16
|
41 |
-
- eval_batch_size: 16
|
42 |
- seed: 42
|
43 |
- gradient_accumulation_steps: 4
|
44 |
- total_train_batch_size: 64
|
|
|
10 |
model-index:
|
11 |
- name: JudgeLLM
|
12 |
results: []
|
13 |
+
tags:
|
14 |
+
- text-generation-inference
|
15 |
+
- question-answering
|
16 |
+
---
|
17 |
+
model-index:
|
18 |
+
- name: Wonder-Griffin/JudgeLLM
|
19 |
+
results:
|
20 |
+
- task:
|
21 |
+
type: text-classification
|
22 |
+
name: Text Classification
|
23 |
+
modality: nlp
|
24 |
+
subtasks:
|
25 |
+
- type: acceptability-classification
|
26 |
+
name: Acceptability Classification
|
27 |
+
- type: sentiment-analysis
|
28 |
+
name: Sentiment Analysis
|
29 |
+
metrics:
|
30 |
+
- name: Accuracy
|
31 |
+
type: accuracy
|
32 |
+
value: 0.95
|
33 |
+
- task:
|
34 |
+
type: summarization
|
35 |
+
name: Summarization
|
36 |
+
modality: nlp
|
37 |
+
metrics:
|
38 |
+
- name: Rouge-L
|
39 |
+
type: rouge
|
40 |
+
value: 0.8
|
41 |
+
inference:
|
42 |
+
parameters:
|
43 |
+
aggregation_strategy: "simple"
|
44 |
+
top_k: 10
|
45 |
+
top_p: 0.9
|
46 |
+
temperature: 0.7
|
47 |
+
max_new_tokens: 50
|
48 |
+
do_sample: true
|
49 |
+
guidance_scale: 7.5
|
50 |
+
num_inference_steps: 50
|
51 |
+
example_inputs:
|
52 |
+
- text: "Sample input text for text classification"
|
53 |
+
table:
|
54 |
+
headers: ["Column 1", "Column 2"]
|
55 |
+
rows:
|
56 |
+
- ["Data 1", "Data 2"]
|
57 |
+
src: "path/to/asset"
|
58 |
+
prompt: "Generate an image with the following prompt..."
|
59 |
+
candidate_labels: ["positive", "negative"]
|
60 |
+
multi_class: true
|
61 |
+
messages:
|
62 |
+
- role: user
|
63 |
+
content: "What is the weather like today?"
|
64 |
+
- role: assistant
|
65 |
+
content: "The weather is sunny with a chance of rain."
|
66 |
+
---
|
67 |
---
|
|
|
|
|
|
|
|
|
68 |
# JudgeLLM
|
69 |
|
70 |
+
This model is a fine-tuned version of [Wonder-Griffin/Judge-GPT2](https://huggingface.co/Wonder-Griffin/Judge-GPT2) on
|
|
|
71 |
## Model description
|
72 |
|
73 |
More information needed
|
|
|
87 |
The following hyperparameters were used during training:
|
88 |
- learning_rate: 5e-05
|
89 |
- train_batch_size: 16
|
|
|
90 |
- seed: 42
|
91 |
- gradient_accumulation_steps: 4
|
92 |
- total_train_batch_size: 64
|