code output sender and execution instructions docs

Author: MikeBirdTech

docs/guides/profiles.mdx

@@ -4,13 +4,38 @@ title: Profiles
 
 Profiles are a powerful way to customize your instance of Open Interpreter.
 
-Profiles can be accessed with `interpreter --profiles`
+Everything from the model to the context window to the message templates can be configured in a profile. This allows you to save multiple variations of Open Interpreter to optimize your specific use-cases.
 
-Profiles can be set with `interpreter --profile name`
+You can access Profiles with `interpreter --profiles`. This will open up the directory where all of your profiles are stored.
 
-### Helpful fields
+To apply a Profile to a session of Open Interpreter, you can run `interpreter --profile <name>`
+
+# Example Profile
+
+```Python
+from interpreter import interpreter
+
+interpreter.os = True
+interpreter.llm.supports_vision = True
+# interpreter.shrink_images = True # Faster but less accurate
+
+interpreter.llm.model = "gpt-4-vision-preview"
+
+interpreter.llm.supports_functions = False
+interpreter.llm.context_window = 110000
+interpreter.llm.max_tokens = 4096
+interpreter.auto_run = True
+interpreter.force_task_completion = True
+```
+
+## Helpful fields for local models
+
+Local models benefit from more coersion and guidance. This can impact the conversational nature of Open Interpreter when the user needs to add extra context to messages. These fields allow templates to be applied to messages to improve the steering of the language model while maintaining the natural conversation.
 
 `interpreter.user_message_template` allows users to have their message wrapped in a template. This can be helpful steering a language model to a desired behaviour without needing the user to add extra context to their message.
+
 `interpreter.always_apply_user_message_template` has all user messages to be wrapped in the template. If False, only the last User message will be wrapped.
+
 `interpreter.code_output_template` wraps the output from the computer after code is run. This can help with nudging the language model to continue working or to explain outputs.
+
 `interpreter.empty_code_output_template` is the message that is sent to the language model if code execution results in no output.

docs/settings/all-settings.mdx

@@ -239,6 +239,22 @@ llm:
 
 </CodeGroup>
 
+### Execution Instructions
+
+If `llm.supports_functions` is `False`, this value will be added to the system message. This parameter tells language models how to execute code. This can be set to an empty string or to `False` if you don't want to tell the LLM how to do this.
+
+<CodeGroup>
+
+````python Python
+interpreter.llm.execution_instructions = "To execute code on the user's machine, write a markdown code block. Specify the language after the ```. You will receive the output. Use any programming language."
+````
+
+````python Profile
+interpreter.llm.execution_instructions = "To execute code on the user's machine, write a markdown code block. Specify the language after the ```. You will receive the output. Use any programming language."
+````
+
+</CodeGroup>
+
 ### LLM Supports Vision
 
 Inform Open Interpreter that the language model you're using supports vision. Defaults to `False`.
@@ -702,6 +718,22 @@ interpreter.empty_code_output_template = "The code above was executed on my mach
 
 </CodeGroup>
 
+### Code Output Sender
+
+This field determines whether the computer / code output messages are sent as the assistant or as the user. The default is user.
+
+<CodeGroup>
+
+```python Python
+interpreter.code_output_sender = "user"
+```
+
+```python Profile
+interpreter.code_output_sender = "assistant"
+```
+
+</CodeGroup>
+
 # Computer
 
 The `computer` object in `interpreter.computer` is a virtual computer that the AI controls. Its primary interface/function is to execute code and return the output in real-time.
@@ -773,3 +805,4 @@ computer.import_computer_api: True
 ```
 
 </CodeGroup>
+````