mirror of
https://github.com/langchain-ai/langsmith-cookbook.git
synced 2026-07-01 08:12:02 -04:00
Update Optimization Links
This commit is contained in:
William FH
committed by
William Fu-Hinthorn
William Fu-Hinthorn
parent
def3eec995
commit
05979983ed
+2
-1
@@ -1,3 +1,4 @@
|
||||
[submodule "optimization/tweet-critic"]
|
||||
path = optimization/tweet-critic
|
||||
url = git@github.com:langchain-ai/tweet-critic.git
|
||||
url = https://github.com/langchain-ai/tweet-critic.git
|
||||
branch = main
|
||||
|
||||
@@ -103,7 +103,7 @@ Use LangSmith to help optimize your LLM systems, so they can continuously learn
|
||||
- [Prompt Bootstrapping](./optimization/assisted-prompt-bootstrapping/assisted-prompt-engineering.ipynb): Optimize your prompt over a set of examples by incorporating human feedback and an LLM prompt optimizer. Works by rewriting an optimized system prompt based on feedback.
|
||||
- [Prompt Bootstrapping for style transfer: Elvis-Bot](./optimization/assisted-prompt-bootstrapping/elvis-bot.ipynb): Extend prompt bootstrapping to generate outputs in the style of a specific persona. This notebook demonstrates how to create an "Elvis-bot" that mimics the tweet style of @omarsar0 by iteratively refining a prompt using Claude's exceptional prompt engineering capabilities and feedback collected through LangSmith's annotation queue.
|
||||
- [Automated Few-shot Prompt Bootstrapping](./optimization/bootstrap-fewshot/bootstrap-few-shot.ipynb): Automatically curate the most informative few-shot examples based on performance metrics, removing the need for manual example engineering. Applied to an entailment task on the SCONE dataset.
|
||||
- [Iterative Prompt Optimization](./optimization/tweet-critic/README.md): Streamlit app demonstrating real-time prompt optimization based on user feedback and dialog, leveraging few-shot learning and a separate "optimizer" model to dynamically improve a tweet-generating system.
|
||||
- [Iterative Prompt Optimization](https://github.com/langchain-ai/tweet-critic): Streamlit app demonstrating real-time prompt optimization based on user feedback and dialog, leveraging few-shot learning and a separate "optimizer" model to dynamically improve a tweet-generating system.
|
||||
- [Online Few-shot Examples](./testing-examples/movie-demo/optimization.ipynb) Configure online evaluators to add good examples to a dataset. Review, then use them as few-shot examples to boost performance.
|
||||
|
||||
## Exporting data for fine-tuning
|
||||
|
||||
@@ -8,5 +8,5 @@ Use LangSmith to help optimize your LLM systems, so they can continuously learn
|
||||
|
||||
- [Prompt Bootstrapping](./assisted-prompt-bootstrapping/assisted-prompt-engineering.ipynb): Optimize your prompt over a set of examples by incorporating human feedback and an LLM prompt optimizer. Works by rewriting an optimized system prompt.
|
||||
- [Prompt Bootstrapping for style transfer: Elvis-Bot](./assisted-prompt-bootstrapping/elvis-bot.ipynb): Extend prompt bootstrapping to generate outputs in the style of a specific persona. This notebook demonstrates how to create an "Elvis-bot" that mimics the tweet style of @omarsar0 by iteratively refining a prompt using Claude's exceptional prompt engineering capabilities and feedback collected through LangSmith's annotation queue.
|
||||
- [Iterative Prompt Optimization](./tweet-critic/README.md): Streamlit app demonstrating real-time prompt optimization based on user feedback and dialog, leveraging few-shot learning and a separate "optimizer" model to dynamically improve a tweet-generating system.
|
||||
- [Iterative Prompt Optimization](https://github.com/langchain-ai/tweet-critic): Streamlit app demonstrating real-time prompt optimization based on user feedback and dialog, leveraging few-shot learning and a separate "optimizer" model to dynamically improve a tweet-generating system.
|
||||
- [Automated Few-shot Prompt Bootstrapping](./bootstrap-fewshot/bootstrap-few-shot.ipynb): Automatically curate the most informative few-shot examples based on performance metrics, removing the need for manual example engineering. Applied to an entailment task on the SCONE dataset.
|
||||
@@ -15,7 +15,7 @@
|
||||
"\n",
|
||||
"LangSmith makes this this whole flow very easy. Let's give it a whirl!\n",
|
||||
"\n",
|
||||
"This example is based on [@alexalbert's example Claude workflow](https://x.com/alexalbert__/status/1767258557039378511?s=20)."
|
||||
"This example is based on [@alexalbert's example Claude workflow](https://twitter.com/alexalbert__/status/1767258557039378511?s=20)."
|
||||
]
|
||||
},
|
||||
{
|
||||
|
||||
@@ -26,7 +26,7 @@
|
||||
"\n",
|
||||
"Enter Claude3, which has [very strong prompt engineering](https://twitter.com/alexalbert__/status/1767258557039378511?s=20) capabilities.\n",
|
||||
"\n",
|
||||
"[`@alexalbert__`]((https://twitter.com/alexalbert__/status/1767258557039378511?s=20)) presented a nice flow for doing this:\n",
|
||||
"[`@alexalbert__`](https://twitter.com/alexalbert__/status/1767258557039378511?s=20) presented a nice flow for doing this:\n",
|
||||
"\n",
|
||||
"```\n",
|
||||
"1. I write an initial prompt for a task.\n",
|
||||
|
||||
Submodule optimization/tweet-critic updated: 78196caee9...1b2ab09745
Reference in New Issue
Block a user