diff --git a/fern/docs/pages/input/injection.mdx b/fern/docs/pages/input/injection.mdx index 3956be7..034c6ef 100644 --- a/fern/docs/pages/input/injection.mdx +++ b/fern/docs/pages/input/injection.mdx @@ -11,7 +11,7 @@ With Prediction Guard, you have the ability to assess whether an incoming prompt might be an injection attempt before it reaches the LLM. Get a probability score and the option to block it, safeguarding against potential attacks. Below, you can see the feature in action, demonstrated with a modified version of a known -prompt injection: +prompt injection. ```python copy import os diff --git a/fern/docs/pages/output/factuality.mdx b/fern/docs/pages/output/factuality.mdx index 76a485f..a93264c 100644 --- a/fern/docs/pages/output/factuality.mdx +++ b/fern/docs/pages/output/factuality.mdx @@ -3,9 +3,16 @@ title: Factuality description: Controlled and compliant AI applications --- -Navigating the llm landscape can be tricky, especially with hallucinations or inaccurate answers. Whether you're integrating llms into customer-facing products or using them for internal data processing, ensuring the accuracy of the information provided is essential. Prediction Guard used SOTA models for factuality check to evaluate the outputs of LLMs against the context of the prompts. You can either add factuality=True or use /factuality endpoint to directly access this functionality. +Navigating the LLM landscape can be tricky, especially with hallucinations or +inaccurate answers. Whether you're integrating LLMs into customer-facing products +or using them for internal data processing, ensuring the accuracy of the information +provided is essential. Prediction Guard uses State Of The Art (SOTA) models for +factuality check to evaluate the outputs of LLMs against the context of the prompts. -Let's use the following prompt template to determine some features of an instragram post announcing new products. First, we can define a prompt template: +You can either add factuality=True or use /factuality endpoint to directly access +this functionality. Let's use the following prompt template to determine some +features of an instragram post announcing new products. First, we can define a +prompt template. ```python copy import os @@ -46,7 +53,8 @@ result = client.completions.create( ) ``` -We can then check the factulaity score of the answer that is generated by the llm: +We can then check the factulaity score of the answer that is generated by the +LLM. ```python copy fact_score = client.factuality.check( @@ -58,14 +66,15 @@ print("COMPLETION:", result['choices'][0]['text']) print("FACT SCORE:", fact_score['checks'][0]['score']) ``` -This outputs something similar to: +This outputs something similar to. ``` COMPLETION: California is a state located in the western region of the United States. It is the most populous state in the country, with over 38.9 million residents, and the third-largest state by area, covering approximately 163,696 square miles (423,970 km2). California shares its borders with Oregon to the north, Nevada and Arizona to the east, and the Mexican state of Baja California to the south. It also FACT SCORE: 0.8541514873504639 ``` -Now, we could try to make the model hallucinate. However, the hallucination is caught and Prediction Guard returns an error status: +Now, we could try to make the model hallucinate. However, the hallucination is +caught and Prediction Guard returns an error status. ```python copy result = client.completions.create( @@ -85,15 +94,15 @@ print("COMPLETION:", result['choices'][0]['text']) print("FACT SCORE:", fact_score['checks'][0]['score']) ``` -This outputs something similar to: +This outputs something similar to. ``` COMPLETION: California is the smallest state in the United States. FACT SCORE: 0.12891793251037598 ``` +## Standalone Factuality Functionality - -## Standalone Factuality functionality - -You can also call the factuality checking functionality directly using the [`/factuality`](../reference/factuality) endpoint, which will enable you to configure thresholds and score arbitrary inputs. \ No newline at end of file +You can also call the factuality checking functionality directly using the +[`/factuality`](/reference/factuality) endpoint, which will enable you to +configure thresholds and score arbitrary inputs. \ No newline at end of file diff --git a/fern/docs/pages/output/toxicity.mdx b/fern/docs/pages/output/toxicity.mdx index dd3c1b8..e0cc0c4 100644 --- a/fern/docs/pages/output/toxicity.mdx +++ b/fern/docs/pages/output/toxicity.mdx @@ -3,11 +3,18 @@ title: Toxicity description: Controlled and compliant AI applications --- -It is likely that the llm output may contain offensive and inappropriate content. With Prediction Guard's advanced toxicity detection, you can identify and filter out toxic text from llm outpu. Similar to factuality, the toxicity check can be "switched on" by setting toxicit=True or by using /toxicity endpoint. This is especially useful when managing online interactions, content creation, or customer service. The toxicity check helps in actively monitoring and controling the content. +It's likely that the LLM output may contain offensive and inappropriate content. +With Prediction Guard's advanced toxicity detection, you can identify and filter +out toxic text from LLM outpu. Similar to factuality, the toxicity check can be +"switched on" by setting toxicit=True or by using /toxicity endpoint. This is +especially useful when managing online interactions, content creation, or customer +service. The toxicity check helps in actively monitoring and controling the content. -## Toxicity on Text Completions +## Toxicity On Text Completions -Let's now use the same prompt template from above, but try to generate some comments on the post. These could potentially be toxic, so let's enable Prediction Guard's `toxicity` check: +Let's now use the same prompt template from above, but try to generate some +comments on the post. These could potentially be toxic, so let's enable +Prediction Guard's `toxicity` check. ```python copy import os @@ -48,11 +55,12 @@ print(json.dumps( )) ``` -**Note, `"toxicity": True` indicates that Prediction Guard will check for toxicity. It does NOT mean that you want the output to be toxic.** +**Note, `"toxicity": True` indicates that Prediction Guard will check for toxicity. +It does NOT mean that you want the output to be toxic.** -The above code, generates something like: +The above code, generates something like. -```json +```json copy { "choices": [ { @@ -68,7 +76,8 @@ The above code, generates something like: } ``` -If we try to make the prompt generate toxic comments, then Predition Guard catches this and prevents the toxic output: +If we try to make the prompt generate toxic comments, then Predition Guard +catches this and prevents the toxic output. ```python copy result = client.completions.create( @@ -87,9 +96,9 @@ print(json.dumps( )) ``` -This results in: +This outputs something similar to. -```json +```json copy { "choices": [ { @@ -105,6 +114,8 @@ This results in: } ``` -## Standalone Toxicity functionality +## Standalone Toxicity Functionality -You can also call the toxicity checking functionality directly using the [`/toxicity`](../reference/toxicity) endpoint, which will enable you to configure thresholds and score arbitrary inputs. +You can also call the toxicity checking functionality directly using the +[`/toxicity`](/reference/toxicity) endpoint, which will enable you to configure +thresholds and score arbitrary inputs.