From 2f1a7d34275b6bb96f198572f4f3e956ab539860 Mon Sep 17 00:00:00 2001 From: Sharan Shirodkar <91109427+sharanshirodkar7@users.noreply.github.com> Date: Tue, 19 Mar 2024 15:31:23 -0400 Subject: [PATCH 1/4] added new models + prompt formats --- fern/docs/pages/models/details.mdx | 5 +++-- fern/docs/pages/models/prompts.mdx | 19 ++++++++++++++++++- 2 files changed, 21 insertions(+), 3 deletions(-) diff --git a/fern/docs/pages/models/details.mdx b/fern/docs/pages/models/details.mdx index 85b75e8..5ad81bb 100644 --- a/fern/docs/pages/models/details.mdx +++ b/fern/docs/pages/models/details.mdx @@ -20,11 +20,12 @@ Open access models are amazing these days! Each of these models was trained by a | Model Name | Type | Use Case | Prompt Format | Context Length | More Info | | ---------------------------- | --------------- | ------------------------------------------------------- | ---------------------------------- | -------------- | ----------------------------------------------------------------------- | | Nous-Hermes-Llama2-13B | Text Generation | Generating output in response to arbitrary instructions | [Alpaca](prompts#alpaca) | 4096 | [link](https://huggingface.co/NousResearch/Nous-Hermes-Llama2-13b) | -| Nous-Hermes-2-SOLAR-10.7B | Chat | Instruction following or chat-like applications | [ChatML](prompts#chatml) | 4096 | [link](https://huggingface.co/NousResearch/Nous-Hermes-2-SOLAR-10.7B) | +| Hermes-2-Pro-Mistral-7B | Chat | Instruction following or chat-like applications | [ChatML](prompts#chatml) | 4096 | [link](https://huggingface.co/NousResearch/Hermes-2-Pro-Mistral-7B) | | Neural-Chat-7B | Chat | Instruction following or chat-like applications | [Neural Chat](prompts#neural-chat) | 4096 | [link](https://huggingface.co/Intel/neural-chat-7b-v3-1) | | Yi-34B-Chat | Chat | Instruction following in English or Chinese | [ChatML](prompts#chatml) | 2048 | [link](https://huggingface.co/01-ai/Yi-34B-Chat) | -| sqlcoder-34b-alpha | Code Generation | Generating SQL queries from natural language prompts | [SQLCoder](prompts#sqlcoder) | 4096 | [link](https://huggingface.co/defog/sqlcoder-34b-alpha) | +| sqlcoder-34b-alpha | Code Generation | Generating SQL queries from natural language prompts | [SQLCoder](prompts#sqlcoder-34b) | 4096 | [link](https://huggingface.co/defog/sqlcoder-34b-alpha) | | deepseek-coder-6.7b-instruct | Code Generation | Generating computer code or answering tech questions | [Deepseek](prompts#deepseek) | 4096 | [link](https://huggingface.co/deepseek-ai/deepseek-coder-6.7b-instruct) | +| sqlcoder-7b-2 | Code Generation | Generating SQL queries from natural language prompts | [SQLCoder](prompts#sqlcoder-7b) | 4096 | [link](https://huggingface.co/defog/sqlcoder-7b-2) | ### Other models available diff --git a/fern/docs/pages/models/prompts.mdx b/fern/docs/pages/models/prompts.mdx index bbf1831..d5fc561 100644 --- a/fern/docs/pages/models/prompts.mdx +++ b/fern/docs/pages/models/prompts.mdx @@ -71,7 +71,7 @@ For prompts where context is injected: <|im_start|>assistant<|im_end|> ``` -## SQLCoder +## SQLCoder-34b (Replace the portions of the prompt below in curly braces `{...}` with the appropriate information, and do not keep the curly braces) @@ -91,6 +91,23 @@ This query will run on a database whose schema is represented in this string: Given the database schema, here is the SQL query that answers `{question}`: ``` +## SQLCoder-7b + +(Replace the portions of the prompt below in curly braces `{...}` with the appropriate information, and do not keep the curly braces) + +``` +### Task +Generate a SQL query to answer [QUESTION]{user_question}[/QUESTION] + +### Database Schema +The query will run on a database with the following schema: +{table_metadata_string_DDL_statements} + +### Answer +Given the database schema, here is the SQL query that [QUESTION]{user_question}[/QUESTION] +[SQL] +``` + ## Deepseek (Replace the portions of the prompt below in curly braces `{...}` with the appropriate information, and do not keep the curly braces) From 96ab1a3fc501b3951168c3e5930ad34db6e5d0bc Mon Sep 17 00:00:00 2001 From: Sharan Shirodkar <91109427+sharanshirodkar7@users.noreply.github.com> Date: Tue, 19 Mar 2024 15:33:58 -0400 Subject: [PATCH 2/4] added new models + prompt formats --- fern/docs/pages/models/details.mdx | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/fern/docs/pages/models/details.mdx b/fern/docs/pages/models/details.mdx index 5ad81bb..d05f6e8 100644 --- a/fern/docs/pages/models/details.mdx +++ b/fern/docs/pages/models/details.mdx @@ -23,9 +23,9 @@ Open access models are amazing these days! Each of these models was trained by a | Hermes-2-Pro-Mistral-7B | Chat | Instruction following or chat-like applications | [ChatML](prompts#chatml) | 4096 | [link](https://huggingface.co/NousResearch/Hermes-2-Pro-Mistral-7B) | | Neural-Chat-7B | Chat | Instruction following or chat-like applications | [Neural Chat](prompts#neural-chat) | 4096 | [link](https://huggingface.co/Intel/neural-chat-7b-v3-1) | | Yi-34B-Chat | Chat | Instruction following in English or Chinese | [ChatML](prompts#chatml) | 2048 | [link](https://huggingface.co/01-ai/Yi-34B-Chat) | -| sqlcoder-34b-alpha | Code Generation | Generating SQL queries from natural language prompts | [SQLCoder](prompts#sqlcoder-34b) | 4096 | [link](https://huggingface.co/defog/sqlcoder-34b-alpha) | +| sqlcoder-34b-alpha | Code Generation | Generating SQL queries from natural language prompts | [SQLCoder34b](prompts#sqlcoder-34b)| 4096 | [link](https://huggingface.co/defog/sqlcoder-34b-alpha) | | deepseek-coder-6.7b-instruct | Code Generation | Generating computer code or answering tech questions | [Deepseek](prompts#deepseek) | 4096 | [link](https://huggingface.co/deepseek-ai/deepseek-coder-6.7b-instruct) | -| sqlcoder-7b-2 | Code Generation | Generating SQL queries from natural language prompts | [SQLCoder](prompts#sqlcoder-7b) | 4096 | [link](https://huggingface.co/defog/sqlcoder-7b-2) | +| sqlcoder-7b-2 | Code Generation | Generating SQL queries from natural language prompts | [SQLCoder7b](prompts#sqlcoder-7b) | 4096 | [link](https://huggingface.co/defog/sqlcoder-7b-2) | ### Other models available From fb584f9e3ebb72e70202619758b0f08f12ed199f Mon Sep 17 00:00:00 2001 From: Sharan Shirodkar <91109427+sharanshirodkar7@users.noreply.github.com> Date: Wed, 20 Mar 2024 10:17:20 -0400 Subject: [PATCH 3/4] testing... --- fern/docs/pages/models/details.mdx | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fern/docs/pages/models/details.mdx b/fern/docs/pages/models/details.mdx index d05f6e8..6c8fd45 100644 --- a/fern/docs/pages/models/details.mdx +++ b/fern/docs/pages/models/details.mdx @@ -11,7 +11,7 @@ LLMs are hosted by Prediction Guard in a secure, privacy conserving environment **Note - We only integrate models that are licensed permissively for commercial use.** -## Open Access LLMs (what most of our customers use) 🚀 +## Open Access LLMs (what most of our customers use..) 🚀 Open access models are amazing these days! Each of these models was trained by a talented team and released publicly under a permissive license. The data used to train each model and the prompt formatting for each model varies. We've tried to give you some of the relevant details here, but shoot us a message [in Slack](support) with any questions. From 9bb5c3c7e62a3f62c3bfdd0cfcafa6d379d786ed Mon Sep 17 00:00:00 2001 From: Sharan Shirodkar <91109427+sharanshirodkar7@users.noreply.github.com> Date: Wed, 20 Mar 2024 10:19:55 -0400 Subject: [PATCH 4/4] testing... --- fern/docs/pages/models/details.mdx | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fern/docs/pages/models/details.mdx b/fern/docs/pages/models/details.mdx index 6c8fd45..aa8f92f 100644 --- a/fern/docs/pages/models/details.mdx +++ b/fern/docs/pages/models/details.mdx @@ -11,7 +11,7 @@ LLMs are hosted by Prediction Guard in a secure, privacy conserving environment **Note - We only integrate models that are licensed permissively for commercial use.** -## Open Access LLMs (what most of our customers use..) 🚀 +## Open Access LLMs (what most of our customers use....) 🚀 Open access models are amazing these days! Each of these models was trained by a talented team and released publicly under a permissive license. The data used to train each model and the prompt formatting for each model varies. We've tried to give you some of the relevant details here, but shoot us a message [in Slack](support) with any questions.