File size: 6,372 Bytes
f527676
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
19ada47
f527676
872b774
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
f527676
19ada47
f527676
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
7b8690f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
f527676
7b8690f
 
 
 
 
 
 
 
 
 
 
 
f527676
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
# AUTOGENERATED! DO NOT EDIT! File to edit: app.ipynb.

# %% auto 0
__all__ = ['title', 'description', 'query_chat_api', 'inference_chat']

# %% app.ipynb 0
import gradio as gr
import requests
import json
import requests

# %% app.ipynb 1
def query_chat_api(
    model_id,
    inputs,
    temperature,
    top_p
):
    API_URL = f"https://api-inference.huggingface.co/models/{model_id}"
    headers = {"Authorization": "Bearer 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", "x-wait-for-model": "1"}

    payload = {
        "inputs": inputs,
        "parameters": {
            "temperature": temperature,
            "top_p": top_p,
            "do_sample": True,
            "max_length": 512,
        },
    }

    response = requests.post(API_URL, json=payload, headers=headers)

    if response.status_code == 200:
        return response.json()
    else:
        return "Error: " + response.text


# %% app.ipynb 4
def inference_chat(
    model_id,
    prompt_template,
    text_input,
    temperature,
    top_p,
    history=[],
):
    with open(f"prompt_templates/{prompt_template}.json", "r") as f:
        prompt_template = json.load(f)

    history.append(text_input)
    inputs = prompt_template["prompt"].format(human_input=text_input)

    output = query_chat_api(model_id, inputs, temperature, top_p)
    history.append(" " + output[0]["generated_text"])

    chat = [
        (history[i], history[i + 1]) for i in range(0, len(history) - 1, 2)
    ]  # convert to tuples of list

    return {chatbot: chat, state: history}


# %% app.ipynb 12
title = """<h1 align="center">Chatty Language Models</h1>"""
description = """Language models can be conditioned to act like dialogue agents through a conversational prompt that typically takes the form:

```
User: <utterance>
Assistant: <utterance>
User: <utterance>
Assistant: <utterance>
...
```

In this app, you can explore the outputs of several language models conditioned on different conversational prompts. The models are trained on different datasets and have different objectives, so they will have different personalities and strengths.

So far, the following prompts are available:

* `langchain_default`: The default prompt used in the [LangChain library](https://github.com/hwchase17/langchain/blob/bc53c928fc1b221d0038b839d111039d31729def/langchain/chains/conversation/prompt.py#L4). Around 67 tokens long.
* `openai_chatgpt`: The prompt used in the OpenAI ChatGPT model. Around 261 tokens long.
* `deepmind_sparrow`: The prompt used in the DeepMind Sparrow model (Table 7 of [their paper](https://arxiv.org/abs/2209.14375)). Around 880 tokens long.
* `deepmind_gopher`: The prompt used in the DeepMind Gopher model (Table A30 of [their paper](https://arxiv.org/abs/2112.11446)). Around 791 tokens long.
* `anthropic_hhh`: The prompt used in the [Anthropic HHH models](https://gist.github.com/jareddk/2509330f8ef3d787fc5aaac67aab5f11#file-hhh_prompt-txt). A whopping 6,341 tokens long!

As you can see, most of these prompts exceed the maximum context size of models like Flan-T5, so an error usually means the Inference API has timed out.
"""

# %% app.ipynb 13
with gr.Blocks(
    css="""
    .message.svelte-w6rprc.svelte-w6rprc.svelte-w6rprc {font-size: 20px; margin-top: 20px}
    #component-21 > div.wrap.svelte-w6rprc {height: 600px;}
    """
) as iface:
    state = gr.State([])

    gr.Markdown(title)
    gr.Markdown(description)

    with gr.Row():
        with gr.Column(scale=1):
            model_id = gr.Dropdown(
                choices=["google/flan-t5-xl"],
                value="google/flan-t5-xl",
                label="Model",
                interactive=True,
            )
            prompt_template = gr.Dropdown(
                choices=[
                    "langchain_default",
                    "openai_chatgpt",
                    "deepmind_sparrow",
                    "deepmind_gopher",
                    "anthropic_hhh",
                ],
                value="langchain_default",
                label="Prompt Template",
                interactive=True,
            )
            temperature = gr.Slider(
                minimum=0.5,
                maximum=3.0,
                value=1.0,
                step=0.1,
                interactive=True,
                label="Temperature",
            )

            top_p = gr.Slider(
                minimum=-0,
                maximum=1.0,
                value=0.95,
                step=0.05,
                interactive=True,
                label="Top-p (nucleus sampling)",
            )

        with gr.Column(scale=1.8):
            with gr.Row():
                with gr.Column(
                    scale=1.5,
                ):
                    chatbot = gr.Chatbot(
                        label="Chat Output",
                    )

                with gr.Column():
                    chat_input = gr.Textbox(lines=1, label="Chat Input")
                    chat_input.submit(
                        inference_chat,
                        [
                            model_id,
                            prompt_template,
                            chat_input,
                            temperature,
                            top_p,
                            state,
                        ],
                        [chatbot, state],
                    )

                    with gr.Row():
                        clear_button = gr.Button(value="Clear", interactive=True)
                        clear_button.click(
                            lambda: ("", [], []),
                            [],
                            [chat_input, chatbot, state],
                            queue=False,
                        )

                        submit_button = gr.Button(
                            value="Submit", interactive=True, variant="primary"
                        )
                        submit_button.click(
                            inference_chat,
                            [
                                model_id,
                                prompt_template,
                                chat_input,
                                temperature,
                                top_p,
                                state,
                            ],
                            [chatbot, state],
                        )
iface.launch()