docs(debugging.md): document new feature

Closes #13814

Author: krrishdholakia

docs/my-website/docs/proxy/debugging.md

@@ -11,65 +11,103 @@ The proxy also supports json logs. [See here](#json-logs)
 
 **via cli**
 
-```bash
+```bash showLineNumbers
 $ litellm --debug
 ```
 
 **via env**
 
-```python
+```python showLineNumbers
 os.environ["LITELLM_LOG"] = "INFO"
 ```
 
 ## `detailed debug`
 
 **via cli**
 
-```bash
+```bash showLineNumbers
 $ litellm --detailed_debug
 ```
 
 **via env**
 
-```python
+```python showLineNumbers
 os.environ["LITELLM_LOG"] = "DEBUG"
 ```
 
 ### Debug Logs 
 
 Run the proxy with `--detailed_debug` to view detailed debug logs
-```shell
+```shell showLineNumbers
 litellm --config /path/to/config.yaml --detailed_debug
 ```
 
 When making requests you should see the POST request sent by LiteLLM to the LLM on the Terminal output
-```shell
+```shell showLineNumbers
 POST Request Sent from LiteLLM:
 curl -X POST \
 https://api.openai.com/v1/chat/completions \
 -H 'content-type: application/json' -H 'Authorization: Bearer sk-qnWGUIW9****************************************' \
 -d '{"model": "gpt-3.5-turbo", "messages": [{"role": "user", "content": "this is a test request, write a short poem"}]}'
 ```
 
+## Debug single request
+
+Pass in `litellm_request_debug=True` in the request body
+
+```bash showLineNumbers
+curl -L -X POST 'http://0.0.0.0:4000/chat/completions' \
+-H 'Content-Type: application/json' \
+-H 'Authorization: Bearer sk-1234' \
+-d '{ 
+    "model":"fake-openai-endpoint",
+    "messages": [{"role": "user","content": "How many r in the word strawberry?"}],
+    "litellm_request_debug": true
+}'
+```
+
+This will emit the raw request sent by LiteLLM to the API Provider and raw response received from the API Provider for **just** this request in the logs. 
+
+
+```bash showLineNumbers
+INFO:     Uvicorn running on http://0.0.0.0:4000 (Press CTRL+C to quit)
+20:14:06 - LiteLLM:WARNING: litellm_logging.py:938 - 
+
+POST Request Sent from LiteLLM:
+curl -X POST \
+https://exampleopenaiendpoint-production.up.railway.app/chat/completions \
+-H 'Authorization: Be****ey' -H 'Content-Type: application/json' \
+-d '{'model': 'fake', 'messages': [{'role': 'user', 'content': 'How many r in the word strawberry?'}], 'stream': False}'
+
+
+20:14:06 - LiteLLM:WARNING: litellm_logging.py:1015 - RAW RESPONSE:
+{"id":"chatcmpl-817fc08f0d6c451485d571dab39b26a1","object":"chat.completion","created":1677652288,"model":"gpt-3.5-turbo-0301","system_fingerprint":"fp_44709d6fcb","choices":[{"index":0,"message":{"role":"assistant","content":"\n\nHello there, how may I assist you today?"},"logprobs":null,"finish_reason":"stop"}],"usage":{"prompt_tokens":9,"completion_tokens":12,"total_tokens":21}}
+
+
+INFO:     127.0.0.1:56155 - "POST /chat/completions HTTP/1.1" 200 OK
+
+```
+
+
 ## JSON LOGS
 
 Set `JSON_LOGS="True"` in your env:
 
-```bash
+```bash showLineNumbers
 export JSON_LOGS="True"
 ```
 **OR**
 
 Set `json_logs: true` in your yaml: 
 
-```yaml
+```yaml showLineNumbers
 litellm_settings:
     json_logs: true
 ```
 
 Start proxy 
 
-```bash
+```bash showLineNumbers
 $ litellm
 ```
 
@@ -80,7 +118,7 @@ The proxy will now all logs in json format.
 Turn off fastapi's default 'INFO' logs 
 
 1. Turn on 'json logs' 
-```yaml
+```yaml showLineNumbers
 litellm_settings:
     json_logs: true
 ```
@@ -89,20 +127,20 @@ litellm_settings:
 
 Only get logs if an error occurs. 
 
-```bash
+```bash showLineNumbers
 LITELLM_LOG="ERROR"
 ```
 
 3. Start proxy 
 
 
-```bash
+```bash showLineNumbers
 $ litellm
 ```
 
 Expected Output: 
 
-```bash
+```bash showLineNumbers
 # no info statements
 ```
 
@@ -119,14 +157,14 @@ This can be caused due to all your models hitting rate limit errors, causing the
 How to control this? 
 - Adjust the cooldown time
 
-```yaml
+```yaml showLineNumbers
 router_settings:
     cooldown_time: 0 # 👈 KEY CHANGE
 ```
 
 - Disable Cooldowns [NOT RECOMMENDED]
 
-```yaml
+```yaml showLineNumbers
 router_settings:
     disable_cooldowns: True
 ```