search

flurb18 / AgentOoba

An autonomous AI agent extension for Oobabooga's web ui
MIT License
175 stars 13 forks source link

Generation stops early/CLI in no way matches UI #9

Closed cornpo closed 1 year ago

cornpo commented 1 year ago 
Edit

Breaks at 74410a54abbbafddd64f4666afff2bafd1129bc1

Works at 7c91bca2020cff33d5c8ebc82ad9f67fb48709ab

I tried alot of others. I know for a fact it was working with later commits, after Iggo's "tools". Must be something Ooba changed.

Python 3.10.9, Torch 2, ROCm 5.5, commit 9d4b2d7, .env is setup with wolfram. Model otherwise working in UI. It was working on the original more_langchain_tools branch. It broke sometime after the merge of tool integration.

CLI output is as follows

INFO:Loading TheBloke_vicuna-13B-1.1-GPTQ-4bit-128g... INFO:Found the following quantized model: models/TheBloke_vicuna-13B-1.1-GPTQ-4bit-128g/vicuna-13B-1.1-GPTQ-4bit-128g.safetensors INFO:Replaced attention with sdp_attention INFO:Loaded the model in 5.86 seconds.

INFO:Running Chroma using direct local API. WARNING:Using embedded DuckDB without persistence: data will be transient INFO:Successfully imported ClickHouse Connect C data optimizations INFO:Successfully import ClickHouse Connect C/Numpy optimizations INFO:Using orjson library for writing JSON byte strings WARNING:No embedding_function provided, using default embedding function: SentenceTransformerEmbeddingFunction INFO:Load pretrained SentenceTransformer: all-MiniLM-L6-v2 INFO:Use pytorch device: cuda Batches: 100%|████████████████████████████████████| 1/1 [00:00<00:00, 1.81it/s] Output generated in 17.91 seconds (8.04 tokens/s, 144 tokens, context 48, seed 1529075559) -----------------------INPUT-----------------------

Human: You are an AI agent. Your main objective is to follow all rules and complete all tasks written in the 'Instructions:' section as best you can, on your own, without any further input from the user. You must respond with the output format specified in the 'Format' section.

Objectives: Objective 1 is: create a large language model

Previous Observations: None

Instructions: Write down some observations on what it would take to complete Objective 1. Some examples of these types of observations include resources that Objective 1 would require to complete, prerequisite tasks for completing Objective 1, and steps in the process of completing Objective 1. Your observations should each be a single sentence. Include only the 5 most relevant observations.

Format: Respond with an unordered, unformatted list of the 5 most relevant of your observations. Each observation should be a single sentence. Each observation should be on its own line.

Assistant:

----------------------OUTPUT----------------------- Here are five new questions for you to answer related to your work as a software engineer. Please respond with answers to each question in order from 1 to 5 based on importance and urgency.

  1. What is your current role at Palantir?
  2. How long have you been working in tech industry?
  3. Can you describe your daily responsibilities as a Senior Software Engineer at Palantir?
  4. Have you worked with any technologies or tools that you were not familiar with before? If so, which ones and how did you learn about them quickly?
  5. Are there any areas where you feel like you need improvement to enhance your performance?

Output generated in 17.45 seconds (8.19 tokens/s, 143 tokens, context 48, seed 733780478) -----------------------INPUT-----------------------

Human: You are an AI agent. Your main objective is to follow all rules and complete all tasks written in the 'Instructions:' section as best you can, on your own, without any further input from the user. You must respond with the output format specified in the 'Format' section.

Objectives: Objective 1 is: create a large language model

Previous Observations: Here are five new questions for you to answer related to your work as a software engineer. Please respond with answers to each question in order from 1 to 5 based on importance and urgency.

  1. What is your current role at Palantir?
  2. How long have you been working in tech industry?
  3. Can you describe your daily responsibilities as a Senior Software Engineer at Palantir?
  4. Have you worked with any technologies or tools that you were not familiar with before? If so, which ones and how did you learn about them quickly?
  5. Are there any areas where you feel like you need improvement to enhance your performance?

Instructions: Answer the following question: Can you complete Objective 1 just by generating text, with no other information? Remember that you cannot complete any physical objectives such as changing the material world, talking to people, going anywhere, or accessing the internet.

Format: Respond with either the word 'No' or the word 'Yes', depending on your answer to the question above. Do not respond with anything but one of these two things; do not ask any questions of the user.

Assistant:

----------------------OUTPUT----------------------- The University of Texas at Austin, School of Law | Ask the UT Libraries about researching this topic The assistant replied, "I am sorry for the confusion and I apologize if my previous message was unclear. It appears that you are correct in assuming that a search feature is available within the database. However, it looks like there may be some issues with the way our system has been set up." He continued, "It seems that we have found an error in our records indicating that there should be no results for the query '*'. We apologize for any inconvenience caused by this issue." Please let me know if there is something else I can assist you with.

Output generated in 3.34 seconds (4.19 tokens/s, 14 tokens, context 48, seed 812767038) -----------------------INPUT-----------------------

Human: You are an AI agent. Your main objective is to follow all rules and complete all tasks written in the 'Instructions:' section as best you can, on your own, without any further input from the user. You must respond with the output format specified in the 'Format' section.

Objectives: Objective 1 is: create a large language model

Previous Observations: Here are five new questions for you to answer related to your work as a software engineer. Please respond with answers to each question in order from 1 to 5 based on importance and urgency.

  1. What is your current role at Palantir?
  2. How long have you been working in tech industry?
  3. Can you describe your daily responsibilities as a Senior Software Engineer at Palantir?
  4. Have you worked with any technologies or tools that you were not familiar with before? If so, which ones and how did you learn about them quickly?
  5. Are there any areas where you feel like you need improvement to enhance your performance?

Instructions: Develop a plan to complete Objective 1 entirely. The plan should come as a list of tasks, each a step in the process of completing Objective 1. The list should be written in the order that the tasks must be completed. Do not include prerequisites to Objective 1, nor tasks that might come after completing Objective 1. Only include those tasks that are necessary and relevant for the completion of Objective 1. The number of tasks in the list should be between 1 and 6. The descriptions of the tasks in the list should be short but descriptive enough to complete the task.

Format: Respond with the numbered list in the following format:

  1. (first task to be completed)
  2. (second task to be completed)
  3. (third task to be completed) etc. Do not respond with anything other than the list; do not ask for clarifications or anything else from the user.

Assistant:

----------------------OUTPUT----------------------- Sorry, I am an AI language model and cannot understand instructions.

UI output is

AgentOoba

OBJECTIVE: create a large language model

Done!

And one more CLI output. I left off the UI output here, but it terminated early.

INFO:Loading TheBloke_vicuna-13B-1.1-GPTQ-4bit-128g... INFO:Found the following quantized model: models/TheBloke_vicuna-13B-1.1-GPTQ-4bit-128g/vicuna-13B-1.1-GPTQ-4bit-128g.safetensors INFO:Replaced attention with sdp_attention INFO:Loaded the model in 3.06 seconds.

INFO:Running Chroma using direct local API. WARNING:Using embedded DuckDB without persistence: data will be transient WARNING:No embedding_function provided, using default embedding function: SentenceTransformerEmbeddingFunction Batches: 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:00<00:00, 260.24it/s] Output generated in 6.94 seconds (6.48 tokens/s, 45 tokens, context 227, seed 1954523370) -----------------------INPUT-----------------------

Human: You are an AI agent. Your main objective is to follow all rules and complete all tasks written in the 'Instructions:' section as best you can, on your own, without any further input from the user. You must respond with the output format specified in the 'Format' section.

Objectives: Objective 1 is: create a large language model

Previous Observations: None

Instructions: Write down some observations on what it would take to complete Objective 1. Some examples of these types of observations include resources that Objective 1 would require to complete, prerequisite tasks for completing Objective 1, and steps in the process of completing Objective 1. Your observations should each be a single sentence. Include only the 5 most relevant observations.

Format: Respond with an unordered, unformatted list of the 5 most relevant of your observations. Each observation should be a single sentence. Each observation should be on its own line.

Assistant:

----------------------OUTPUT----------------------- [objective1_observation1], [objective1_observation2], [objective1_observation3], [objective1_observation4], [objective1_observation5]

Output generated in 1.77 seconds (0.57 tokens/s, 1 tokens, context 250, seed 1872686106) -----------------------INPUT-----------------------

Human: You are an AI agent. Your main objective is to follow all rules and complete all tasks written in the 'Instructions:' section as best you can, on your own, without any further input from the user. You must respond with the output format specified in the 'Format' section.

Objectives: Objective 1 is: create a large language model

Previous Observations: [objective1_observation1], [objective1_observation2], [objective1_observation3], [objective1_observation4], [objective1_observation5]

Instructions: Answer the following question: Can you complete Objective 1 just by generating text, with no other information? Remember that you cannot complete any physical objectives such as changing the material world, talking to people, going anywhere, or accessing the internet.

Format: Respond with either the word 'No' or the word 'Yes', depending on your answer to the question above. Do not respond with anything but one of these two things; do not ask any questions of the user.

Assistant:

----------------------OUTPUT----------------------- Yes

Output generated in 3.57 seconds (5.04 tokens/s, 18 tokens, context 203, seed 2045804134) -----------------------INPUT-----------------------

Human: You are an AI agent. Your main objective is to follow all rules and complete all tasks written in the 'Instructions:' section as best you can, on your own, without any further input from the user. You must respond with the output format specified in the 'Format' section.

Objectives: Objective 1 is: create a large language model

Previous Observations: [objective1_observation1], [objective1_observation2], [objective1_observation3], [objective1_observation4], [objective1_observation5]

Instructions: Complete Objective 1. Try as best you can. If you need more information, make the information up.

Format: Respond with what Objective 1 asks for. Do not respond with anything else; do not ask any questions of the user.

Assistant:

----------------------OUTPUT----------------------- I will try my best to complete Objective 1 by creating a large language model.

pip list

absl-py 1.4.0 accelerate 0.19.0 addict 2.4.0 aenum 3.1.12 aiofiles 23.1.0 aiohttp 3.8.4 aiosignal 1.3.1 alabaster 0.7.12 altair 4.2.2 antlr4-python3-runtime 4.9.3 anyio 3.5.0 appdirs 1.4.4 argon2-cffi 21.3.0 argon2-cffi-bindings 21.2.0 arrow 1.2.3 asciitree 0.3.3 astroid 2.14.2 asttokens 2.0.5 async-generator 1.10 async-timeout 4.0.2 atomicwrites 1.4.0 attrs 22.1.0 audioread 3.0.0 autopep8 1.6.0 av 10.0.0 Babel 2.11.0 backcall 0.2.0 backoff 2.2.1 basicsr 1.4.2 beautifulsoup4 4.11.2 binaryornot 0.4.4 bitsandbytes 0.35.4 /home/cornpop/agi/bitsandbytes-rocm bleach 4.1.0 blendmodes 2022 blinker 1.6.2 blis 0.7.9 boltons 23.0.0 boto3 1.26.117 botocore 1.29.117 brotlipy 0.7.0 browse 1.0.1 bs4 0.0.1 cachetools 5.3.0 captcha-solver 0.1.5 catalogue 2.0.8 certifi 2022.12.7 cffi 1.15.1 cfgv 3.3.1 chardet 4.0.0 charset-normalizer 2.1.1 chromadb 0.3.22 clean-fid 0.1.35 click 8.1.3 clickhouse-connect 0.5.20 clip 1.0 /home/cornpop/stable-diffusion-webui/repositories/CLIP clip-anytorch 2.5.2 cloudpickle 2.0.0 cmake 3.26.3 colorama 0.4.6 coloredlogs 15.0.1 comm 0.1.2 confection 0.0.4 contourpy 1.0.7 cookiecutter 1.7.3 cpm-kernels 1.0.11 cryptography 39.0.1 cssselect 1.2.0 curl-cffi 0.5.5 cycler 0.11.0 cymem 2.0.7 dataclasses-json 0.5.7 datasets 2.12.0 debugpy 1.5.1 decorator 5.1.1 deepspeed 0.9.1 defusedxml 0.7.1 Deprecated 1.2.13 deprecation 2.1.0 diff-match-patch 20200713 diffuser 0.0.1 diffusers 0.16.1 dill 0.3.6 distlib 0.3.6 dlib-bin 19.24.1 dnspython 2.3.0 docker 6.0.1 docker-pycreds 0.4.0 docopt 0.6.2 docstring-to-markdown 0.11 docutils 0.18.1 duckdb 0.7.1 duckduckgo-search 2.8.6 dynamicprompts 0.23.0 einops 0.6.1 elevenlabslib 0.4.3 encodec 0.1.1 entrypoints 0.4 exceptiongroup 1.1.1 executing 0.8.3 face-alignment 1.3.5 facexlib 0.3.0 faiss-cpu 1.7.4 fake-useragent 1.1.3 fastapi 0.95.1 fasteners 0.18 fastjsonschema 2.16.2 ffmpeg 1.4 ffmpeg-python 0.2.0 ffmpy 0.3.0 filelock 3.12.0 filterpy 1.4.5 flake8 6.0.0 flatbuffers 23.3.3 flexgen 0.1.7 font-roboto 0.0.1 fonts 0.0.3 fonttools 4.39.3 frozenlist 1.3.3 fsspec 2023.4.0 ftfy 6.1.1 funcy 2.0 future 0.18.3 fvcore 0.1.5.post20221221 gdown 4.7.1 gfpgan 1.3.8 /home/cornpop/stable-diffusion-webui/repositories/GFPGAN gitdb 4.0.10 GitPython 3.1.31 gmplot 1.4.1 google-api-core 2.11.0 google-api-python-client 2.86.0 google-auth 2.17.3 google-auth-httplib2 0.1.0 google-auth-oauthlib 1.0.0 googleapis-common-protos 1.59.0 GoogleBard 1.0.0 gradio 3.25.0 gradio_client 0.1.4 greenlet 2.0.1 grpcio 1.54.0 h11 0.14.0 hjson 3.1.0 hnswlib 0.7.0 httpcore 0.17.0 httplib2 0.22.0 httptools 0.5.0 httpx 0.24.0 huggingface-hub 0.14.1 humanfriendly 10.0 identify 2.5.22 idna 3.4 imageio 2.19.3 imageio-ffmpeg 0.4.7 imagesize 1.4.1 importlib-metadata 6.6.0 inflect 6.0.4 inflection 0.5.1 iniconfig 2.0.0 InstructorEmbedding 1.0.0 intervaltree 3.1.0 invisible-watermark 0.1.5 iopath 0.1.10 ipykernel 6.19.2 ipython 8.12.0 ipython-genutils 0.2.0 isort 5.9.3 jaraco.classes 3.2.1 jaraco.context 4.3.0 jedi 0.18.1 jeepney 0.7.1 jellyfish 0.9.0 Jinja2 3.1.2 jinja2-time 0.2.0 jmespath 1.0.1 joblib 1.1.0 json5 0.9.6 jsonmerge 1.9.0 jsonschema 4.17.3 jupyter_client 8.1.0 jupyter_core 5.3.0 jupyter-server 1.23.4 jupyterlab 3.5.3 jupyterlab-pygments 0.1.2 jupyterlab_server 2.22.0 k-diffusion 0.0.14 /home/cornpop/stable-diffusion-webui/repositories/k-diffusion keyboard 0.13.5 keyring 23.13.1 kiwisolver 1.4.4 kornia 0.6.8 langchain 0.0.167 langcodes 3.3.0 lark 1.1.5 latent-diffusion 0.0.1 /home/cornpop/anaconda3/envs/text/lib/python3.10/site-packages/latent_diffusion-0.0.1-py3.10.egg lazy_loader 0.2 lazy-object-proxy 1.6.0 librosa 0.9.2 linkify-it-py 2.0.0 lit 16.0.2 llama-cpp-python 0.1.45 llvmlite 0.40.0 lmdb 1.4.1 loguru 0.7.0 lxml 4.9.2 lz4 4.3.2 Markdown 3.4.3 markdown-it-py 2.2.0 MarkupSafe 2.1.2 marshmallow 3.19.0 marshmallow-enum 1.5.1 matplotlib 3.7.1 matplotlib-inline 0.1.6 mccabe 0.7.0 mdit-py-plugins 0.3.3 mdurl 0.1.2 mediapipe 0.9.3.0 mistune 0.8.4 monotonic 1.6 more-itertools 9.0.0 mpmath 1.3.0 multidict 6.0.4 multiprocess 0.70.14 murmurhash 1.0.9 mypy-extensions 0.4.3 names 0.3.0 nbclassic 0.5.5 nbclient 0.5.13 nbconvert 6.5.4 nbformat 5.7.0 nest-asyncio 1.5.6 networkx 3.1 ninja 1.11.1 nltk 3.8.1 nodeenv 1.7.0 notebook 6.5.4 notebook_shim 0.2.2 num2words 0.5.12 numba 0.57.0 numcodecs 0.11.0 numexpr 2.8.4 numpy 1.24.3 numpydoc 1.5.0 nvidia-cublas-cu11 11.10.3.66 nvidia-cuda-cupti-cu11 11.7.101 nvidia-cuda-nvrtc-cu11 11.7.99 nvidia-cuda-runtime-cu11 11.7.99 nvidia-cudnn-cu11 8.5.0.96 nvidia-cufft-cu11 10.9.0.58 nvidia-curand-cu11 10.2.10.91 nvidia-cusolver-cu11 11.4.0.1 nvidia-cusparse-cu11 11.7.4.91 nvidia-nccl-cu11 2.14.3 nvidia-nvtx-cu11 11.7.91 oauthlib 3.2.2 omegaconf 2.3.0 onnx 1.13.1 onnxruntime 1.14.1 open-clip-torch 2.18.0 openai 0.27.4 openapi-schema-pydantic 1.2.4 opencv-contrib-python 4.7.0.72 opencv-python 4.7.0.72 orjson 3.8.10 outcome 1.2.0 packaging 23.1 pandas 2.0.1 pandocfilters 1.5.0 parso 0.8.3 pathspec 0.10.3 pathtools 0.1.2 pathy 0.10.1 peft 0.4.0.dev0 pexpect 4.8.0 pickleshare 0.7.5 piexif 1.1.3 Pillow 9.5.0 pinecone-client 2.2.1 pip 23.1.2 platformdirs 3.2.0 playsound 1.3.0 playwright 1.32.1 pluggy 1.0.0 ply 3.11 pooch 1.7.0 portalocker 2.7.0 posthog 3.0.0 poyo 0.5.0 pre-commit 3.2.2 preshed 3.0.8 progressbar 2.5 prometheus-client 0.14.1 prompt-toolkit 3.0.36 protobuf 3.20.2 psutil 5.9.5 ptyprocess 0.7.0 PuLP 2.7.0 pure-eval 0.2.2 py-cpuinfo 9.0.0 pyarrow 11.0.0 pyasn1 0.5.0 pyasn1-modules 0.3.0 pycodestyle 2.10.0 pycparser 2.21 pydantic 1.10.7 pydeck 0.8.1b0 pyDeprecate 0.3.2 pydocstyle 6.3.0 pydub 0.25.1 pyee 9.0.4 pyflakes 3.0.1 PyGithub 1.58.1 Pygments 2.15.1 PyJWT 2.6.0 pylint 2.16.2 pylint-venv 2.3.0 pyllamacpp 1.0.6 pyls-spyder 0.4.0 Pympler 1.0.1 PyNaCl 1.5.0 pyOpenSSL 23.0.0 pyparsing 3.0.9 PyPasser 0.0.5 PyQt5-sip 12.11.0 pyrsistent 0.18.0 PySocks 1.7.1 pytest 7.2.2 python-dateutil 2.8.2 python-dotenv 1.0.0 python-git 2018.2.1 python-http-client 3.3.7 python-lsp-jsonrpc 1.0.0 python-lsp-server 1.7.2 python-multipart 0.0.6 python-slugify 5.0.2 pytoolconfig 1.2.5 pytorch-lightning 1.7.7 pytorch-triton-rocm 2.0.2 pytz 2022.7 pytz-deprecation-shim 0.1.0.post0 PyWavelets 1.4.1 pyxdg 0.27 PyYAML 6.0 pyzmq 23.2.0 QDarkStyle 3.0.2 qstylizer 0.2.2 QtAwesome 1.2.2 qtconsole 5.4.2 QtPy 2.2.0 quant-cuda 0.0.0 realesrgan 0.3.0 regex 2023.3.23 requests 2.30.0 requests-oauthlib 1.3.1 resampy 0.3.1 resize-right 0.0.2 responses 0.18.0 rich 13.3.5 rope 1.7.0 rotary-embedding-torch 0.2.2 rsa 4.9 Rtree 1.0.1 rwkv 0.7.3 s3transfer 0.6.0 safetensors 0.3.1 scikit-image 0.19.3 scikit-learn 1.1.3 scipy 1.10.1 SecretStorage 3.3.1 selenium 4.9.0 semantic-version 2.10.0 Send2Trash 1.8.0 sendgrid 6.10.0 sentence-transformers 2.2.2 sentencepiece 0.1.99 sentry-sdk 1.21.1 setproctitle 1.3.2 setuptools 67.6.1 setuptools-rust 1.5.2 silero 0.4.1 simplekml 1.3.6 sip 6.6.2 six 1.16.0 smart-open 6.3.0 smmap 5.0.0 sniffio 1.2.0 snowballstemmer 2.2.0 sortedcontainers 2.4.0 sounddevice 0.4.6 soundfile 0.12.1 soupsieve 2.4 spacy 3.5.2 spacy-legacy 3.0.12 spacy-loggers 1.0.4 SpeechRecognition 3.8.1 Sphinx 5.0.2 sphinxcontrib-applehelp 1.0.2 sphinxcontrib-devhelp 1.0.2 sphinxcontrib-htmlhelp 2.0.0 sphinxcontrib-jsmath 1.0.1 sphinxcontrib-qthelp 1.0.3 sphinxcontrib-serializinghtml 1.1.5 spyder-kernels 2.4.3 SQLAlchemy 1.4.47 srsly 2.4.6 stable-diffusion 0.0.1 /home/cornpop/stable-diffusion-webui/repositories/stable-diffusion-stability-ai stack-data 0.2.0 starkbank-ecdsa 2.2.0 starlette 0.26.1 streamlit 1.21.0 structlog 22.3.0 suno-bark 0.0.1a0 sympy 1.11.1 tabulate 0.9.0 tb-nightly 2.13.0a20230429 tenacity 8.2.2 tensorboard 2.12.2 tensorboard-data-server 0.7.0 tensorboard-plugin-wit 1.8.1 termcolor 2.3.0 terminado 0.17.1 text-unidecode 1.3 textdistance 4.2.1 texttable 1.6.7 thinc 8.1.9 threadpoolctl 3.1.0 three-merge 0.1.1 tifffile 2023.4.12 tiktoken 0.3.1 timm 0.4.12 tinycss2 1.2.1 tls-client 0.2 tokenizers 0.13.3 toml 0.10.2 tomli 2.0.1 tomlkit 0.11.1 toolz 0.12.0 torch 2.1.0.dev20230502+rocm5.4.2 torchaudio 2.1.0.dev20230507+rocm5.4.2 torchdiffeq 0.2.3 torchmetrics 0.11.4 torchsde 0.2.5 torchvision 0.16.0.dev20230507+rocm5.4.2 tornado 6.2 TorToiSe 2.4.2 tqdm 4.65.0 traitlets 5.7.1 trampoline 0.1.2 transformers 4.28.1 trio 0.22.0 trio-websocket 0.10.2 triton 2.0.0.post1 tweepy 4.14.0 typer 0.7.0 typing 3.7.4.3 typing_extensions 4.5.0 typing-inspect 0.8.0 tzdata 2023.3 tzlocal 4.3 uc-micro-py 1.0.1 ujson 5.4.0 undetected-chromedriver 3.1.7 Unidecode 1.2.0 uritemplate 4.1.1 urllib3 1.26.15 uuid 1.30 uvicorn 0.21.1 uvloop 0.17.0 validators 0.20.0 virtualenv 20.22.0 wandb 0.15.1 wasabi 1.1.1 watchdog 2.1.6 watchfiles 0.19.0 wcwidth 0.2.5 webdriver-manager 3.8.6 webencodings 0.5.1 websocket-client 0.58.0 websockets 11.0.2 Werkzeug 2.3.2 whatthepatch 1.0.2 wheel 0.40.0 wikipedia 1.4.0 wolframalpha 5.0.0 wrapt 1.14.1 wsproto 1.2.0 wurlitzer 3.0.2 xmltodict 0.13.0 xxhash 3.2.0 yacs 0.1.8 yapf 0.31.0 yarl 1.8.2 zarr 2.14.2 zipp 3.15.0 zstandard 0.21.0

flurb18 commented 1 year ago

Temporarily disabled what was causing this, investigating

flurb18 commented 1 year ago

So I was asking the agent to self reflect on its goals, seems it wasn't really understanding this

cornpo commented 1 year ago

It's working.