Commit Graph

113 Commits

Author SHA1 Message Date
Rafael dos Santos Silva d692ecc7de
FIX: Disable truncation and padding in all-mpnet-base-v2 tokenizer (#105)
The tokenizer was truncating and padding to 128 tokens, and we try append
new post content until we hit 384 tokens. This was causing the tokenizer
to accept all posts in a topic, wasting CPU and memory.
2023-07-13 21:09:46 -03:00
Rafael dos Santos Silva 703762a7a9
PERF: .find_each instead of .find to save us from memory allocation peaks
also Fix embeddings rake task for new db structure
2023-07-13 18:59:25 -03:00
Roman Rizzi 5f0c617880
REFACTOR: Cohesive narrative for single-chunk summaries. (#103)
Single and multi-chunk summaries end using different prompts for the last summary. This change detects when the summarized content fits in a single chunk and uses a slightly different prompt, which leads to more consistent summary formats.

This PR also moves the chunk-splitting step to the `FoldContent` strategy as preparation for implementing streamed summaries.
2023-07-13 17:05:41 -03:00
David Taylor 48d880d3c8
FIX: Rerender related topics correctly when topic changes (#100)
* FIX: Rerender related topics correctly when topic changes


Co-authored-by: Rafael dos Santos Silva <xfalcox@gmail.com>
2023-07-13 16:34:02 -03:00
Rafael dos Santos Silva 5e3f4e1b78
FEATURE: Embeddings to main db (#99)
* FEATURE: Embeddings to main db

This commit moves our embeddings store from an external configurable PostgreSQL
instance back into the main database. This is done to simplify the setup.

There is a migration that will try to import the external embeddings into
the main DB if it is configured and there are rows.

It removes support from embeddings models that aren't all_mpnet_base_v2 or OpenAI
text_embedding_ada_002. However it will now be easier to add new models.

It also now takes into account:
  - topic title
  - topic category
  - topic tags
  - replies (as much as the model allows)

We introduce an interface so we can eventually support multiple strategies
for handling long topics.

This PR severely damages the semantic search performance, but this is a
temporary until we can get adapt HyDE to make semantic search use the same
embeddings we have for semantic related with good performance.

Here we also have some ground work to add post level embeddings, but this
will be added in a future PR.

Please note that this PR will also block Discourse from booting / updating if 
this plugin is installed and the pgvector extension isn't available on the 
PostgreSQL instance Discourse uses.
2023-07-13 12:41:36 -03:00
Rafael dos Santos Silva 9d10a152b9
FEATURE: Claude 2 for summarization and AIHelper (#101) 2023-07-13 12:32:08 -03:00
dependabot[bot] 1c017c22b0
Build(deps): Bump semver from 6.3.0 to 6.3.1 (#102)
Bumps [semver](https://github.com/npm/node-semver) from 6.3.0 to 6.3.1.
- [Release notes](https://github.com/npm/node-semver/releases)
- [Changelog](https://github.com/npm/node-semver/blob/v6.3.1/CHANGELOG.md)
- [Commits](https://github.com/npm/node-semver/compare/v6.3.0...v6.3.1)

---
updated-dependencies:
- dependency-name: semver
  dependency-type: indirect
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2023-07-13 09:33:34 +02:00
Roman Rizzi fbe1bab980
FIX: typo while updating a section (#98) 2023-06-27 17:57:58 -03:00
Roman Rizzi 1b568f2391
FIX: Claude's max_tookens_to_sample is a required field (#97) 2023-06-27 14:42:33 -03:00
Roman Rizzi 9a79afcdbf
DEV: Better strategies for summarization (#88)
* DEV: Better strategies for summarization

The strategy responsibility needs to be "Given a collection of texts, I know how to summarize them most efficiently, using the minimum amount of requests and maximizing token usage".

There are different token limits for each model, so it all boils down to two different strategies:

Fold all these texts into a single one, doing the summarization in chunks, and then build a summary from those.
Build it by combining texts in a single prompt, and truncate it according to your token limits.

While the latter is less than ideal, we need it for "bart-large-cnn-samsum" and "flan-t5-base-samsum", both with low limits. The rest will rely on folding.

* Expose summarized chunks to users
2023-06-27 12:26:33 -03:00
Sam 9390fba768
FIX: adjust token limits to account for functions (#96)
Reduce maximum replies to 2500 tokens and make them even for both GPT-3.5
and 4

Account for 400+ tokens in function definitions (this was unaccounted for)
2023-06-23 10:02:04 +10:00
Sam f8cabfad6b
FEATURE: Try to hone search so it reduces search terms in subsequent rounds (#95)
Teach via system message that you can reduce search terms to get more
results
2023-06-21 20:07:55 +10:00
Sam a028309cbd
FEATURE: add ai_bot_enabled_chat commands and tune search (#94)
* FEATURE: add ai_bot_enabled_chat commands and tune search

This allows admins to disable/enable GPT command integrations.

Also hones search results which were looping cause the result did not denote
the failure properly (it lost context)

* include more context for google command
include more context for time command

* type
2023-06-21 17:10:30 +10:00
Sam d1ab79e82f
FEATURE: Add Azure cognitive service support (#93)
The new site settings:

ai_openai_gpt35_url : distribution for GPT 16k
ai_openai_gpt4_url: distribution for GPT 4
ai_openai_embeddings_url: distribution for ada2

If untouched we will simply use OpenAI endpoints.

Azure requires 1 URL per model, OpenAI allows a single URL to serve multiple models. Hence the new settings.
2023-06-21 10:39:51 +10:00
Sam 30778d8af8
FIX: avoid storing corrupt prompts (#92)
```
prompt << build_message(bot_user.username, reply)
```

Would store a "cooked" prompt which is invalid, instead just store the raw
values which are later passed to build_message

Additionally:

1. Disable summary command which needs honing
2. Stop storing decorations (searched for X) in prompt which leads to straying
3. Ship username directly to model, avoiding "user: content" in prompts. This
 was causing GPT to stray
2023-06-20 15:44:03 +10:00
Sam 70c158cae1
FEATURE: add full bot support for GPT 3.5 (#87)
Given latest GPT 3.5 16k which is both better steered and supports functions
we can now support rich bot integration.

Clunky system message based steering is removed and instead we use the
function framework provided by Open AI
2023-06-20 08:45:31 +10:00
Rafael dos Santos Silva e457c687ca
FIX: OpenAI Tokenizer was failing to truncate mid emojis (#91)
* FIX: OpenAI Tokenizer was failing to truncate mid emojis

* Update spec/shared/tokenizer.rb

Co-authored-by: Joffrey JAFFEUX <j.jaffeux@gmail.com>

---------

Co-authored-by: Joffrey JAFFEUX <j.jaffeux@gmail.com>
2023-06-16 15:15:36 -03:00
Roman Rizzi 9e901dbfbf
FIX: Serialize channel title for DMs (#90) 2023-06-16 14:37:16 -03:00
David Taylor 68bc80e19d
DEV: Fix Zeitwerk reloading error (#89)
Using `.append` can lead to a 'two routes with the same name' error
2023-06-16 11:32:54 +01:00
Rafael dos Santos Silva 8742535024
FEATURE: Allow using large context OpenAI models for summarization (#86) 2023-06-13 15:23:48 -03:00
Roman Rizzi 3364fec425
DEV: Remove the summarization feature (#83)
* DEV: Remove the summarization feature

Instead, we'll register summarization implementations for OpenAI, Anthropic, and Discourse AI using the API defined in discourse/discourse#21813.

Core and chat will implement features on top of these implementations instead of this plugin extending them.

* Register instances that contain the model, requiring less site settings
2023-06-13 14:32:26 -03:00
Sam 081231a6eb
FIX: support multiple command executions (#85)
Previous to this change we were chaining stuff too late and would execute
commands serially leading to very unexpected results

This corrects this and allows us to run stuff like:

> Search google 3/4 times on various permutations of
QUERY and answer this question.

We limit at 5 commands to ensure there are not pathological user cases
where you lean on the LLM to flood us with results.
2023-06-06 07:09:33 +10:00
Sam 840968630e
FEATURE: disable smart commands on Claude and GPT 3.5 (#84)
For the time being smart commands only work consistently on GPT 4.
Avoid using any smart commands on the earlier models.

Additionally adds better error handling to Claude which sometimes streams
partial json and slightly tunes the search command.
2023-06-01 09:10:33 +10:00
Sam 96d521198b
FIX: missing localization (#81)
blog.start_gpt_chat -> was on my blog

This also slightly tunes the search prompt to support filtering by oldest
and try a tiny bit harder to guide GPT 3.5 which is a bit of a losing battle

Co-authored-by: Krzysztof Kotlarek <kotlarek.krzysztof@gmail.com>
2023-05-25 11:05:02 +10:00
Rafael dos Santos Silva cfc6e388df
FIX: Ensure embeddings database outages are handled gracefully (#80)
The rails_failover middleware will intercept all `PG::ConnectionBad` errors and put the cluster into readonly mode. It does not have any handling for multiple databases. Therefore, an issue with the embeddings database was taking the whole cluster into readonly.

This commit fixes the issue by rescuing `PG::Error` from all AI database accesses, and re-raises errors with a different class. It also adds a spec to ensure that an embeddings database outage does not affect the functionality of the topics/show route.

Co-authored-by: David Taylor <david@taylorhq.com>
2023-05-23 22:57:52 +01:00
Rafael dos Santos Silva b213fe7f94
FIX: Give up trying to reuse the DB connection and rely on pgbouncer (#79) 2023-05-23 15:12:59 -03:00
Roman Rizzi c582e3b848
DEV: Fix toxicity test (#78) 2023-05-23 11:02:11 -03:00
Sam d85b503ed4
FIX: guide GPT 3.5 better (#77)
* FIX: guide GPT 3.5 better

This limits search results to 10 cause we were blowing the whole token
budget on search results, additionally it includes a quick exchange at
the start of a session to try and guide GPT 3.5 to follow instructions

Sadly GPT 3.5 drifts off very quickly but this does improve stuff a bit.

It also attempts to correct some issues with anthropic, though it still is
surprisingly hard to ground

* add status:public, this is a bit of a hack but ensures that we can search
for any filter provided

* fix specs
2023-05-23 23:08:17 +10:00
Sam b82fc1e692
FIX: ensure we only attempt embedding once every 15 minutes (#76)
This also heavily reduced log noise and ensures our exception handling is
more surgical.
2023-05-23 10:43:24 +10:00
Sam 074d00ca32
FEATURE: improve search prompt (#75)
- We only support searching public topics - make it clear
- Stop using bug/feature, cause is poisons system - these may not exist
- Add after: and before: which are very handy for bounding search results
2023-05-23 07:52:14 +10:00
Sam e0cf7b7d70
FIX: results will be nil for invalid queries (#74)
Previous to this change invalid searches would break the command.
2023-05-22 15:14:26 +10:00
Sam 92fb84e24d
iterate commands (#73)
* FEATURE: introduce a more efficient formatter

Previous formatting style was space inefficient given JSON consumes lots
of tokens, the new format is now used consistently across commands

Also fixes

- search limited to 10
- search breaking on limit: non existent directive

* Slight improvement to summarizer
Stop blowing up context with custom prompts

* ensure we include the guiding message

* correct spec

* langchain style summarizer ...

much more accurate (albeit more expensive)

* lint
2023-05-22 12:09:14 +10:00
Sam d59ed1091b
FEATURE: add support for GPT <-> Forum integration
This change-set connects GPT based chat with the forum it runs on. Allowing it to perform search, lookup tags and categories and summarize topics. 

The integration is currently restricted to public portions of the forum. 

Changes made:

- Do not run ai reply job for small actions
- Improved composable system prompt
- Trivial summarizer for topics
- Image generator 
- Google command for searching via Google
- Corrected trimming of posts raw (was replacing with numbers) 
- Bypass of problem specs

The feature works best with GPT-4


---------

Co-authored-by: Roman Rizzi <rizziromanalejandro@gmail.com>
2023-05-20 17:45:54 +10:00
Rafael dos Santos Silva 262ed4753e
FEATURE: Basic StableDiffusion text2img support (#72) 2023-05-20 09:38:08 +10:00
Roman Rizzi ba65d42940
FIX: Hide summarization button from anons (#70) 2023-05-18 12:34:49 -03:00
David Taylor 261fe13599
DEV: Tweaks for drop-down implementation (#69)
- Remove unused 'toggleAiBotPanel' widget action

- Switch from appEvents to closure actions

- Convert widget definition to native class syntax, so that we can use `@action` decorator. (alternatively, we could have done `{ closePanel: this.hideAiBotPanel.bind(this) }` in `RenderGlimmer`
2023-05-18 09:28:43 +01:00
Sam deb34bb52f
UX: improve drop down menu for enabled bots (#68)
Previously we were not using using HeaderPanel for drop down, which caused
it not to properly act like a header panel.

- Not styled right
- Not hidden when other buttons clicked

Etc...

Header is sadly full of legacy so this is somewhat hacky weaving widgets.
2023-05-18 16:10:08 +10:00
Rafael dos Santos Silva 739b314312
Fixes for embeddings and truncate (#67) 2023-05-18 09:21:28 +10:00
Roman Rizzi 9ae8f86850
UX: Improvements to the AI Bot header shortcut (#66)
- Humanize model names
- Focus composer editor when starting a PM with a bot
- Correctly close the models dropdown when clicking outside
2023-05-17 15:59:48 -03:00
Rafael dos Santos Silva e9ae28f773
FIX: Non instructor OSS embeddings was broken (#65) 2023-05-17 12:10:10 -03:00
Roman Rizzi 5ade5d7404
UX: Add missing label translation (#64) 2023-05-16 16:56:30 -03:00
Roman Rizzi 362f6167d1
FEATURE: Less friction for starting a conversation with an AI bot. (#63)
* FEATURE: Less friction for starting a conversation with an AI bot.

This PR adds a new header icon as a shortcut to start a conversation with one of our AI Bots. After clicking and selecting one from the dropdown menu, we'll open the composer with some fields already filled (recipients and title).

If you leave the title as is, we'll queue a job after five minutes to update it using a bot suggestion.

* Update assets/javascripts/initializers/ai-bot-replies.js

Co-authored-by: Rafael dos Santos Silva <xfalcox@gmail.com>

* Update assets/javascripts/initializers/ai-bot-replies.js

Co-authored-by: Rafael dos Santos Silva <xfalcox@gmail.com>

---------

Co-authored-by: Rafael dos Santos Silva <xfalcox@gmail.com>
2023-05-16 14:38:21 -03:00
Rafael dos Santos Silva 2ed1f874c2
Use correct API signature for instructor embeddings (#62) 2023-05-15 17:18:11 -03:00
Rafael dos Santos Silva 3c9513e754
Refinements to embeddings and tokenizers (#61)
* Refinements to embeddings and tokenizers

* lint

* Truncate with tokenizers for summary

* fix
2023-05-15 15:10:42 -03:00
Sam 93d9d9ea91
FEATURE: hide various api keys in site settings (#59)
Also... ensures that plugin has a link to more info and proper description
2023-05-12 12:54:25 +10:00
Rafael dos Santos Silva 97124b30de
FEATURE: Update summarization token count and add Claude 100k (#58) 2023-05-11 15:35:58 -03:00
Rafael dos Santos Silva 66bf4c74c6
FEATURE: Handle invalid media in NSFW module (#57)
* FEATURE: Handle invalid media in NSFW module

* fix lint
2023-05-11 15:35:39 -03:00
Roman Rizzi 7e3cb0ea16
FEATURE: Multi-model support for the AI Bot module. (#56)
We'll create one bot user for each available model. When listed in the `ai_bot_enabled_chat_bots` setting, they will reply.

This PR lets us use Claude-v1 in stream mode.
2023-05-11 10:03:03 -03:00
Rafael dos Santos Silva e5537d4c77
FEATURE: Allow excluding closed topics from semantic related (#55) 2023-05-09 15:30:50 -03:00
Rafael dos Santos Silva f1133f66a6
Updates to embedding rake tasks (#54)
- Creates embeddings in topic ID order, so it's easier to stop and
restart from where we stopped

- Update index parameters with current best practices
2023-05-09 13:45:16 -03:00