Update results/GenericAgent-Claude-3.5-Sonnet/README.md
Browse files
    	
        results/GenericAgent-Claude-3.5-Sonnet/README.md
    CHANGED
    
    | 
         @@ -1 +1,44 @@ 
     | 
|
| 1 | 
         
            -
            ### Claude 
     | 
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
| 
         | 
|
| 1 | 
         
            +
            ### GenericAgent-Claude-3.5-Sonnet
         
     | 
| 2 | 
         
            +
             
     | 
| 3 | 
         
            +
            This agent is [GenericAgent](https://github.com/ServiceNow/AgentLab/blob/main/src/agentlab/agents/generic_agent/generic_agent.py) from [AgentLab](https://github.com/ServiceNow/AgentLab)
         
     | 
| 4 | 
         
            +
             
     | 
| 5 | 
         
            +
            It uses Claude-3.5-sonnet as a backend, with the following [flags](https://github.com/ServiceNow/AgentLab/blob/main/src/agentlab/agents/generic_agent/tmlr_config.py):
         
     | 
| 6 | 
         
            +
            ```python
         
     | 
| 7 | 
         
            +
            BASE_FLAGS = GenericPromptFlags(
         
     | 
| 8 | 
         
            +
                obs=dp.ObsFlags(
         
     | 
| 9 | 
         
            +
                    use_html=False,
         
     | 
| 10 | 
         
            +
                    use_ax_tree=True,
         
     | 
| 11 | 
         
            +
                    use_focused_element=True,
         
     | 
| 12 | 
         
            +
                    use_error_logs=True,
         
     | 
| 13 | 
         
            +
                    use_history=True,
         
     | 
| 14 | 
         
            +
                    use_past_error_logs=False,
         
     | 
| 15 | 
         
            +
                    use_action_history=True,
         
     | 
| 16 | 
         
            +
                    use_think_history=True,  # gpt-4o config except for this line
         
     | 
| 17 | 
         
            +
                    use_diff=False,
         
     | 
| 18 | 
         
            +
                    html_type="pruned_html",
         
     | 
| 19 | 
         
            +
                    use_screenshot=False,
         
     | 
| 20 | 
         
            +
                    use_som=False,
         
     | 
| 21 | 
         
            +
                    extract_visible_tag=True,
         
     | 
| 22 | 
         
            +
                    extract_clickable_tag=True,
         
     | 
| 23 | 
         
            +
                    extract_coords="False",
         
     | 
| 24 | 
         
            +
                    filter_visible_elements_only=False,
         
     | 
| 25 | 
         
            +
                ),
         
     | 
| 26 | 
         
            +
                action=dp.ActionFlags(
         
     | 
| 27 | 
         
            +
                    multi_actions=False,
         
     | 
| 28 | 
         
            +
                    action_set="bid",
         
     | 
| 29 | 
         
            +
                    long_description=False,
         
     | 
| 30 | 
         
            +
                    individual_examples=False,
         
     | 
| 31 | 
         
            +
                ),
         
     | 
| 32 | 
         
            +
                use_plan=False,
         
     | 
| 33 | 
         
            +
                use_criticise=False,
         
     | 
| 34 | 
         
            +
                use_thinking=True,
         
     | 
| 35 | 
         
            +
                use_memory=False,
         
     | 
| 36 | 
         
            +
                use_concrete_example=True,
         
     | 
| 37 | 
         
            +
                use_abstract_example=True,
         
     | 
| 38 | 
         
            +
                use_hints=True,
         
     | 
| 39 | 
         
            +
                enable_chat=False,
         
     | 
| 40 | 
         
            +
                max_prompt_tokens=40_000,
         
     | 
| 41 | 
         
            +
                be_cautious=True,
         
     | 
| 42 | 
         
            +
                extra_instructions=None,
         
     | 
| 43 | 
         
            +
            )
         
     | 
| 44 | 
         
            +
            ```
         
     |