feat: add HTML rendering for humor datasets

This commit is contained in:
Kirill Igumenshchev 2025-05-18 17:55:59 -07:00 committed by Kirill Igumenshchev (aider)
parent 96043a968f
commit 41cf093415
40 changed files with 2272 additions and 0 deletions

View file

@ -0,0 +1 @@
run-20250518_174835-2wfxhi28/logs/debug-internal.log

View file

@ -0,0 +1 @@
run-20250518_174835-2wfxhi28/logs/debug.log

View file

@ -0,0 +1 @@
run-20250518_174835-2wfxhi28

View file

@ -0,0 +1,78 @@
_wandb:
value:
cli_version: 0.19.11
m: []
python_version: 3.13.2
t:
"1":
- 1
- 11
- 49
- 51
- 55
- 95
"2":
- 1
- 11
- 49
- 51
- 55
- 95
"3":
- 13
- 16
- 23
- 55
"4": 3.13.2
"5": 0.19.11
"6": 4.51.3
"8":
- 5
"12": 0.19.11
"13": linux-x86_64
batch_size:
value: -1
data_path:
value: humor_dataset.jsonl
data_path_to_save_groups:
value: data/humor_1.jsonl
ensure_scores_are_not_same:
value: false
eval_handling:
value: STOP_TRAIN
eval_limit_ratio:
value: 0.5
group_size:
value: 2
include_messages:
value: true
inference_weight:
value: 1
max_batches_offpolicy:
value: 3
max_eval_workers:
value: 16
max_num_workers:
value: -1
max_num_workers_per_node:
value: 8
max_token_length:
value: 2048
min_items_sent_before_logging:
value: 2
num_rollouts_per_group_for_logging:
value: 1
num_rollouts_to_keep:
value: 32
rollout_server_url:
value: http://localhost:8000
steps_per_eval:
value: 100
tokenizer_name:
value: NousResearch/DeepHermes-3-Llama-3-3B-Preview
total_steps:
value: 2
use_wandb:
value: true
wandb_name:
value: null

View file

@ -0,0 +1,189 @@
Starting to process 2 groups...
Processing group 1/2
Traceback (most recent call last):
File "/home/kirill/miniconda3/lib/python3.13/asyncio/runners.py", line 118, in run
return self._loop.run_until_complete(task)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^
File "/home/kirill/miniconda3/lib/python3.13/asyncio/base_events.py", line 725, in run_until_complete
return future.result()
~~~~~~~~~~~~~^^
File "/home/kirill/miniconda3/lib/python3.13/site-packages/atroposlib/envs/base.py", line 959, in process_manager
to_postprocess, _ = await self.collect_trajectories(item)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/kirill/hackathons/atropos-nrs15h/environments/hack0/llm_humor_server.py", line 61, in collect_trajectories
chat_completions = await self.server.chat_completion(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
...<3 lines>...
)
^
File "/home/kirill/miniconda3/lib/python3.13/site-packages/atroposlib/envs/server_handling/server_manager.py", line 176, in chat_completion
return await self.servers[most_available_server].chat_completion(**kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/kirill/miniconda3/lib/python3.13/site-packages/tenacity/asyncio/__init__.py", line 189, in async_wrapped
return await copy(fn, *args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/kirill/miniconda3/lib/python3.13/site-packages/tenacity/asyncio/__init__.py", line 111, in __call__
do = await self.iter(retry_state=retry_state)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/kirill/miniconda3/lib/python3.13/site-packages/tenacity/asyncio/__init__.py", line 153, in iter
result = await action(retry_state)
^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/kirill/miniconda3/lib/python3.13/site-packages/tenacity/_utils.py", line 99, in inner
return call(*args, **kwargs)
File "/home/kirill/miniconda3/lib/python3.13/site-packages/tenacity/__init__.py", line 400, in <lambda>
self._add_action_func(lambda rs: rs.outcome.result())
~~~~~~~~~~~~~~~~~^^
File "/home/kirill/miniconda3/lib/python3.13/concurrent/futures/_base.py", line 449, in result
return self.__get_result()
~~~~~~~~~~~~~~~~~^^
File "/home/kirill/miniconda3/lib/python3.13/concurrent/futures/_base.py", line 401, in __get_result
raise self._exception
File "/home/kirill/miniconda3/lib/python3.13/site-packages/tenacity/asyncio/__init__.py", line 114, in __call__
result = await fn(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/kirill/miniconda3/lib/python3.13/site-packages/atroposlib/envs/server_handling/server_baseline.py", line 279, in chat_completion
ret_data = await self._chat_comp(stat_dict, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/kirill/miniconda3/lib/python3.13/site-packages/tenacity/asyncio/__init__.py", line 189, in async_wrapped
return await copy(fn, *args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/kirill/miniconda3/lib/python3.13/site-packages/tenacity/asyncio/__init__.py", line 111, in __call__
do = await self.iter(retry_state=retry_state)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/kirill/miniconda3/lib/python3.13/site-packages/tenacity/asyncio/__init__.py", line 153, in iter
result = await action(retry_state)
^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/kirill/miniconda3/lib/python3.13/site-packages/tenacity/_utils.py", line 99, in inner
return call(*args, **kwargs)
File "/home/kirill/miniconda3/lib/python3.13/site-packages/tenacity/__init__.py", line 400, in <lambda>
self._add_action_func(lambda rs: rs.outcome.result())
~~~~~~~~~~~~~~~~~^^
File "/home/kirill/miniconda3/lib/python3.13/concurrent/futures/_base.py", line 449, in result
return self.__get_result()
~~~~~~~~~~~~~~~~~^^
File "/home/kirill/miniconda3/lib/python3.13/concurrent/futures/_base.py", line 401, in __get_result
raise self._exception
File "/home/kirill/miniconda3/lib/python3.13/site-packages/tenacity/asyncio/__init__.py", line 114, in __call__
result = await fn(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/kirill/miniconda3/lib/python3.13/site-packages/atroposlib/envs/server_handling/server_baseline.py", line 233, in _chat_comp
completions = await self._chat_completion_wrapper(**kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/kirill/miniconda3/lib/python3.13/site-packages/atroposlib/envs/server_handling/openai_server.py", line 78, in _chat_completion_wrapper
completions = await self.openai.chat.completions.create(**kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/kirill/miniconda3/lib/python3.13/site-packages/openai/resources/chat/completions/completions.py", line 2028, in create
return await self._post(
^^^^^^^^^^^^^^^^^
...<45 lines>...
)
^
File "/home/kirill/miniconda3/lib/python3.13/site-packages/openai/_base_client.py", line 1742, in post
return await self.request(cast_to, opts, stream=stream, stream_cls=stream_cls)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/kirill/miniconda3/lib/python3.13/site-packages/openai/_base_client.py", line 1484, in request
response = await self._client.send(
^^^^^^^^^^^^^^^^^^^^^^^^
...<3 lines>...
)
^
File "/home/kirill/miniconda3/lib/python3.13/site-packages/httpx/_client.py", line 1629, in send
response = await self._send_handling_auth(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
...<4 lines>...
)
^
File "/home/kirill/miniconda3/lib/python3.13/site-packages/httpx/_client.py", line 1657, in _send_handling_auth
response = await self._send_handling_redirects(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
...<3 lines>...
)
^
File "/home/kirill/miniconda3/lib/python3.13/site-packages/httpx/_client.py", line 1694, in _send_handling_redirects
response = await self._send_single_request(request)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/kirill/miniconda3/lib/python3.13/site-packages/httpx/_client.py", line 1730, in _send_single_request
response = await transport.handle_async_request(request)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/kirill/miniconda3/lib/python3.13/site-packages/httpx/_transports/default.py", line 394, in handle_async_request
resp = await self._pool.handle_async_request(req)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/kirill/miniconda3/lib/python3.13/site-packages/httpcore/_async/connection_pool.py", line 256, in handle_async_request
raise exc from None
File "/home/kirill/miniconda3/lib/python3.13/site-packages/httpcore/_async/connection_pool.py", line 236, in handle_async_request
response = await connection.handle_async_request(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
pool_request.request
^^^^^^^^^^^^^^^^^^^^
)
^
File "/home/kirill/miniconda3/lib/python3.13/site-packages/httpcore/_async/connection.py", line 103, in handle_async_request
return await self._connection.handle_async_request(request)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/kirill/miniconda3/lib/python3.13/site-packages/httpcore/_async/http11.py", line 136, in handle_async_request
raise exc
File "/home/kirill/miniconda3/lib/python3.13/site-packages/httpcore/_async/http11.py", line 106, in handle_async_request
) = await self._receive_response_headers(**kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/kirill/miniconda3/lib/python3.13/site-packages/httpcore/_async/http11.py", line 177, in _receive_response_headers
event = await self._receive_event(timeout=timeout)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/kirill/miniconda3/lib/python3.13/site-packages/httpcore/_async/http11.py", line 217, in _receive_event
data = await self._network_stream.read(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
self.READ_NUM_BYTES, timeout=timeout
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
)
^
File "/home/kirill/miniconda3/lib/python3.13/site-packages/httpcore/_backends/anyio.py", line 35, in read
return await self._stream.receive(max_bytes=max_bytes)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/kirill/miniconda3/lib/python3.13/site-packages/anyio/streams/tls.py", line 219, in receive
data = await self._call_sslobject_method(self._ssl_object.read, max_bytes)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/kirill/miniconda3/lib/python3.13/site-packages/anyio/streams/tls.py", line 162, in _call_sslobject_method
data = await self.transport_stream.receive()
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/kirill/miniconda3/lib/python3.13/site-packages/anyio/_backends/_asyncio.py", line 1254, in receive
await self._protocol.read_event.wait()
File "/home/kirill/miniconda3/lib/python3.13/asyncio/locks.py", line 213, in wait
await fut
asyncio.exceptions.CancelledError
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "/home/kirill/hackathons/atropos-nrs15h/environments/hack0/llm_humor_server.py", line 101, in <module>
HumorEnv.cli()
~~~~~~~~~~~~^^
File "/home/kirill/miniconda3/lib/python3.13/site-packages/atroposlib/envs/base.py", line 1013, in cli
run_and_exit(
~~~~~~~~~~~~^
subcommands,
^^^^^^^^^^^^
description=f"CLI for {cls.__name__}",
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
exception_handler=custom_error_handler,
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
)
^
File "/home/kirill/miniconda3/lib/python3.13/site-packages/pydantic_cli/__init__.py", line 644, in run_and_exit
sys.exit(f(_args))
~^^^^^^^
File "/home/kirill/miniconda3/lib/python3.13/site-packages/pydantic_cli/__init__.py", line 523, in f
return _runner(
args,
...<4 lines>...
epilogue_handler,
)
File "/home/kirill/miniconda3/lib/python3.13/site-packages/pydantic_cli/__init__.py", line 326, in _runner
out = cmd.run()
File "/home/kirill/miniconda3/lib/python3.13/site-packages/atroposlib/envs/base.py", line 1382, in run
asyncio.run(env.process_manager())
~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^
File "/home/kirill/miniconda3/lib/python3.13/asyncio/runners.py", line 195, in run
return runner.run(main)
~~~~~~~~~~^^^^^^
File "/home/kirill/miniconda3/lib/python3.13/asyncio/runners.py", line 123, in run
raise KeyboardInterrupt()
KeyboardInterrupt

View file

@ -0,0 +1,178 @@
google-api-python-client==2.169.0
aiofiles==24.1.0
datasets==3.6.0
requests==2.32.3
nvidia-cusolver-cu12==11.7.1.2
babel==2.17.0
grpcio==1.71.0
polars==1.29.0
packaging==24.2
anaconda-anon-usage==0.7.0
nvidia-cufft-cu12==11.3.0.4
wandb==0.19.11
attrs==25.3.0
starlette==0.46.2
six==1.17.0
cloudpickle==3.1.1
numpy==2.2.6
pycparser==2.21
protobuf==5.29.4
cachetools==5.5.2
nvidia-cublas-cu12==12.6.4.1
uvloop==0.21.0
jsonpointer==2.1
psutil==7.0.0
conda==25.3.1
nvidia-nccl-cu12==2.26.2
h11==0.16.0
click-default-group==1.2.4
nvidia-cuda-cupti-cu12==12.6.80
uvicorn==0.34.2
click==8.1.8
xxhash==3.5.0
multiprocess==0.70.16
sentry-sdk==2.28.0
python-dateutil==2.9.0.post0
pyarrow==20.0.0
annotated-types==0.6.0
pydantic_core==2.27.1
pyasn1_modules==0.4.2
pyasn1==0.6.1
llm-cmd-comp==1.1.1
tqdm==4.67.1
pyparsing==3.2.3
wcwidth==0.2.13
llm==0.25
networkx==3.4.2
conda_package_streaming==0.11.0
zstandard==0.23.0
nvidia-nvtx-cu12==12.6.77
python-dotenv==1.1.0
pydantic_cli==9.1.0
tabulate==0.9.0
nvidia-cuda-nvrtc-cu12==12.6.77
frozendict==2.4.2
google-ai-generativelanguage==0.6.15
truststore==0.10.0
Brotli==1.0.9
Markdown==3.8
nvidia-cusparse-cu12==12.5.4.2
llm-groq==0.8
puremagic==1.29
tzlocal==5.3.1
mpmath==1.3.0
python-ulid==3.0.0
sqlite-migrate==0.1b0
jiter==0.9.0
llm-cmd==0.2a0
htmldate==1.9.3
condense-json==0.1.2
dill==0.3.8
atroposlib==0.2.1
math-verify==0.7.0
MarkupSafe==3.0.2
jsonlines==4.0.0
yarl==1.20.0
httptools==0.6.4
jusText==3.0.2
multidict==6.4.3
nvidia-cudnn-cu12==9.5.1.17
Farama-Notifications==0.0.4
aiohappyeyeballs==2.6.1
openai==1.78.0
ruamel.yaml==0.18.10
PyYAML==6.0.2
Pygments==2.19.1
PySocks==1.7.1
boltons==24.1.0
propcache==0.3.1
conda-package-handling==2.4.0
anyio==4.9.0
conda-content-trust==0.2.0
google-api-core==2.25.0rc0
wheel==0.45.1
fsspec==2025.3.0
libmambapy==2.0.5
smmap==5.0.2
markdown-it-py==2.2.0
gitdb==4.0.12
prompt_toolkit==3.0.51
regex==2024.11.6
nvidia-nvjitlink-cu12==12.6.85
proto-plus==1.26.1
google-generativeai==0.8.5
cffi==1.17.1
torch==2.7.0
certifi==2025.4.26
llm-jq==0.1.1
lxml==5.4.0
trafilatura==2.0.0
antlr4-python3-runtime==4.13.2
platformdirs==4.3.7
sqlite-utils==3.38
Jinja2==3.1.6
lxml_html_clean==0.4.2
typing_extensions==4.12.2
huggingface-hub==0.31.2
joblib==1.5.0
triton==3.3.0
rich==13.9.4
nvidia-curand-cu12==10.3.7.77
aiohttp==3.11.18
distro==1.9.0
pip==25.0
urllib3==2.3.0
setproctitle==1.3.6
pandas==2.2.3
nvidia-cufile-cu12==1.11.1.6
watchfiles==1.0.5
jsonpatch==1.33
ruamel.yaml.clib==0.2.12
conda-anaconda-telemetry==0.1.2
sniffio==1.3.1
nvidia-cuda-runtime-cu12==12.6.77
pycosat==0.6.6
conda-anaconda-tos==0.1.3
latex2sympy2_extended==1.10.1
grpcio-status==1.71.0
conda-libmamba-solver==25.4.0
transformers==4.51.3
nltk==3.9.1
sympy==1.14.0
gymnasium==1.1.1
groq==0.24.0
hf_transfer==0.1.9
courlan==1.3.2
google-auth==2.39.0
pytz==2025.2
docker-pycreds==0.4.0
filelock==3.18.0
dateparser==1.2.1
mdurl==0.1.0
GitPython==3.1.44
uritemplate==4.1.1
menuinst==2.2.0
pydantic==2.10.3
httpx==0.28.1
googleapis-common-protos==1.70.0
frozenlist==1.6.0
google-auth-httplib2==0.2.0
aiosignal==1.3.2
websockets==15.0.1
tld==0.13
rsa==4.9.1
httpcore==1.0.9
setuptools==78.1.1
tenacity==9.1.2
httplib2==0.22.0
sqlite-fts4==1.0.3
cryptography==44.0.1
fastapi==0.115.12
archspec==0.2.3
pluggy==1.5.0
idna==3.7
tzdata==2025.2
tokenizers==0.21.1
nvidia-cusparselt-cu12==0.6.3
charset-normalizer==3.4.2
safetensors==0.5.3

View file

@ -0,0 +1,44 @@
{
"os": "Linux-5.15.167.4-microsoft-standard-WSL2-x86_64-with-glibc2.40",
"python": "CPython 3.13.2",
"startedAt": "2025-05-19T00:45:10.647164Z",
"args": [
"process",
"--openai.base_url",
"https://api.x.ai/v1",
"--openai.api_key",
"xai-cjDD6fd8ArfNJgc5Xnz3AjiJNuUxMuaD1rcpTAOR4RRh0SSO8v2KBwjYwUuEI2kFQxQOkUx943keKiUF",
"--openai.model_name",
"grok-3-beta",
"--env.group_size",
"2",
"--env.data_path",
"humor_dataset.jsonl"
],
"program": "/home/kirill/hackathons/atropos-nrs15h/environments/hack0/llm_humor_server.py",
"codePath": "environments/hack0/llm_humor_server.py",
"git": {
"remote": "git@github.com:kirilligum/atropos-nrs15h.git",
"commit": "db1e68d2abd2f64515325da6aac5a0484efd1741"
},
"email": "kirill.igum@gmail.com",
"root": "/home/kirill/hackathons/atropos-nrs15h/environments/hack0",
"host": "koldun",
"executable": "/home/kirill/miniconda3/bin/python",
"codePathLocal": "llm_humor_server.py",
"cpu_count": 7,
"cpu_count_logical": 14,
"disk": {
"/": {
"total": "1081101176832",
"used": "161502097408"
}
},
"memory": {
"total": "16480206848"
},
"cpu": {
"count": 7,
"countLogical": 14
}
}

View file

@ -0,0 +1 @@
{"_wandb":{"runtime":18}}

View file

@ -0,0 +1 @@
/home/kirill/.cache/wandb/logs/core-debug-20250518_174510.log

View file

@ -0,0 +1,15 @@
{"time":"2025-05-18T17:45:10.65168024-07:00","level":"INFO","msg":"stream: starting","core version":"0.19.11","symlink path":"/home/kirill/hackathons/atropos-nrs15h/environments/hack0/wandb/run-20250518_174510-m5eheo08/logs/debug-core.log"}
{"time":"2025-05-18T17:45:11.000138776-07:00","level":"INFO","msg":"created new stream","id":"m5eheo08"}
{"time":"2025-05-18T17:45:11.000177601-07:00","level":"INFO","msg":"stream: started","id":"m5eheo08"}
{"time":"2025-05-18T17:45:11.000219532-07:00","level":"INFO","msg":"writer: Do: started","stream_id":"m5eheo08"}
{"time":"2025-05-18T17:45:11.000293159-07:00","level":"INFO","msg":"sender: started","stream_id":"m5eheo08"}
{"time":"2025-05-18T17:45:11.000319305-07:00","level":"INFO","msg":"handler: started","stream_id":"m5eheo08"}
{"time":"2025-05-18T17:45:11.194295311-07:00","level":"INFO","msg":"Starting system monitor"}
{"time":"2025-05-18T17:45:28.741546685-07:00","level":"INFO","msg":"stream: closing","id":"m5eheo08"}
{"time":"2025-05-18T17:45:28.741574051-07:00","level":"INFO","msg":"Stopping system monitor"}
{"time":"2025-05-18T17:45:28.74159836-07:00","level":"INFO","msg":"Stopped system monitor"}
{"time":"2025-05-18T17:45:29.119087953-07:00","level":"INFO","msg":"fileTransfer: Close: file transfer manager closed"}
{"time":"2025-05-18T17:45:29.241156378-07:00","level":"INFO","msg":"handler: closed","stream_id":"m5eheo08"}
{"time":"2025-05-18T17:45:29.241243031-07:00","level":"INFO","msg":"writer: Close: closed","stream_id":"m5eheo08"}
{"time":"2025-05-18T17:45:29.24127025-07:00","level":"INFO","msg":"sender: closed","stream_id":"m5eheo08"}
{"time":"2025-05-18T17:45:29.241302715-07:00","level":"INFO","msg":"stream: closed","id":"m5eheo08"}

View file

@ -0,0 +1,23 @@
2025-05-18 17:45:10,643 INFO MainThread:67867 [wandb_setup.py:_flush():70] Current SDK version is 0.19.11
2025-05-18 17:45:10,643 INFO MainThread:67867 [wandb_setup.py:_flush():70] Configure stats pid to 67867
2025-05-18 17:45:10,643 INFO MainThread:67867 [wandb_setup.py:_flush():70] Loading settings from /home/kirill/.config/wandb/settings
2025-05-18 17:45:10,643 INFO MainThread:67867 [wandb_setup.py:_flush():70] Loading settings from /home/kirill/hackathons/atropos-nrs15h/environments/hack0/wandb/settings
2025-05-18 17:45:10,643 INFO MainThread:67867 [wandb_setup.py:_flush():70] Loading settings from environment variables
2025-05-18 17:45:10,643 INFO MainThread:67867 [wandb_init.py:setup_run_log_directory():724] Logging user logs to /home/kirill/hackathons/atropos-nrs15h/environments/hack0/wandb/run-20250518_174510-m5eheo08/logs/debug.log
2025-05-18 17:45:10,643 INFO MainThread:67867 [wandb_init.py:setup_run_log_directory():725] Logging internal logs to /home/kirill/hackathons/atropos-nrs15h/environments/hack0/wandb/run-20250518_174510-m5eheo08/logs/debug-internal.log
2025-05-18 17:45:10,644 INFO MainThread:67867 [wandb_init.py:init():852] calling init triggers
2025-05-18 17:45:10,644 INFO MainThread:67867 [wandb_init.py:init():857] wandb.init called with sweep_config: {}
config: {'group_size': 2, 'max_num_workers': -1, 'max_eval_workers': 16, 'max_num_workers_per_node': 8, 'steps_per_eval': 100, 'max_token_length': 2048, 'eval_handling': <EvalHandlingEnum.STOP_TRAIN: 'STOP_TRAIN'>, 'eval_limit_ratio': 0.5, 'inference_weight': 1.0, 'batch_size': -1, 'max_batches_offpolicy': 3, 'tokenizer_name': 'NousResearch/DeepHermes-3-Llama-3-3B-Preview', 'use_wandb': True, 'rollout_server_url': 'http://localhost:8000', 'total_steps': 2, 'wandb_name': None, 'num_rollouts_to_keep': 32, 'num_rollouts_per_group_for_logging': 1, 'ensure_scores_are_not_same': False, 'data_path_to_save_groups': 'data/humor_1.jsonl', 'min_items_sent_before_logging': 2, 'include_messages': True, 'data_path': 'humor_dataset.jsonl', '_wandb': {}}
2025-05-18 17:45:10,644 INFO MainThread:67867 [wandb_init.py:init():893] starting backend
2025-05-18 17:45:10,644 INFO MainThread:67867 [wandb_init.py:init():897] sending inform_init request
2025-05-18 17:45:10,646 INFO MainThread:67867 [backend.py:_multiprocessing_setup():101] multiprocessing start_methods=fork,spawn,forkserver, using: spawn
2025-05-18 17:45:10,647 INFO MainThread:67867 [wandb_init.py:init():907] backend started and connected
2025-05-18 17:45:10,648 INFO MainThread:67867 [wandb_init.py:init():1005] updated telemetry
2025-05-18 17:45:10,652 INFO MainThread:67867 [wandb_init.py:init():1029] communicating run to backend with 90.0 second timeout
2025-05-18 17:45:11,192 INFO MainThread:67867 [wandb_init.py:init():1104] starting run threads in backend
2025-05-18 17:45:11,290 INFO MainThread:67867 [wandb_run.py:_console_start():2573] atexit reg
2025-05-18 17:45:11,291 INFO MainThread:67867 [wandb_run.py:_redirect():2421] redirect: wrap_raw
2025-05-18 17:45:11,291 INFO MainThread:67867 [wandb_run.py:_redirect():2490] Wrapping output streams.
2025-05-18 17:45:11,291 INFO MainThread:67867 [wandb_run.py:_redirect():2513] Redirects installed.
2025-05-18 17:45:11,292 INFO MainThread:67867 [wandb_init.py:init():1150] run started, returning control to user process
2025-05-18 17:45:28,740 INFO MsgRouterThr:67867 [mailbox.py:close():129] [no run ID] Closing mailbox, abandoning 1 handles.

View file

@ -0,0 +1,79 @@
_wandb:
value:
cli_version: 0.19.11
m: []
python_version: 3.13.2
t:
"1":
- 1
- 11
- 49
- 51
- 55
- 95
"2":
- 1
- 11
- 49
- 51
- 55
- 95
"3":
- 13
- 16
- 23
- 55
- 61
"4": 3.13.2
"5": 0.19.11
"6": 4.51.3
"8":
- 5
"12": 0.19.11
"13": linux-x86_64
batch_size:
value: -1
data_path:
value: humor_dataset.jsonl
data_path_to_save_groups:
value: data/humor_2.jsonl
ensure_scores_are_not_same:
value: false
eval_handling:
value: STOP_TRAIN
eval_limit_ratio:
value: 0.5
group_size:
value: 2
include_messages:
value: true
inference_weight:
value: 1
max_batches_offpolicy:
value: 3
max_eval_workers:
value: 16
max_num_workers:
value: -1
max_num_workers_per_node:
value: 8
max_token_length:
value: 2048
min_items_sent_before_logging:
value: 2
num_rollouts_per_group_for_logging:
value: 1
num_rollouts_to_keep:
value: 32
rollout_server_url:
value: http://localhost:8000
steps_per_eval:
value: 100
tokenizer_name:
value: NousResearch/DeepHermes-3-Llama-3-3B-Preview
total_steps:
value: 2
use_wandb:
value: true
wandb_name:
value: null

View file

@ -0,0 +1,9 @@
Starting to process 2 groups...
Processing group 1/2
Wrote scored group to data/humor_2.jsonl
Successfully processed group 1/2
Processing group 2/2
Wrote scored group to data/humor_2.jsonl
Successfully processed group 2/2
Completed processing 2 groups
Successfully generated HTML file: /home/kirill/hackathons/atropos-nrs15h/environments/hack0/data/humor_2.html

View file

@ -0,0 +1,178 @@
google-api-python-client==2.169.0
aiofiles==24.1.0
datasets==3.6.0
requests==2.32.3
nvidia-cusolver-cu12==11.7.1.2
babel==2.17.0
grpcio==1.71.0
polars==1.29.0
packaging==24.2
anaconda-anon-usage==0.7.0
nvidia-cufft-cu12==11.3.0.4
wandb==0.19.11
attrs==25.3.0
starlette==0.46.2
six==1.17.0
cloudpickle==3.1.1
numpy==2.2.6
pycparser==2.21
protobuf==5.29.4
cachetools==5.5.2
nvidia-cublas-cu12==12.6.4.1
uvloop==0.21.0
jsonpointer==2.1
psutil==7.0.0
conda==25.3.1
nvidia-nccl-cu12==2.26.2
h11==0.16.0
click-default-group==1.2.4
nvidia-cuda-cupti-cu12==12.6.80
uvicorn==0.34.2
click==8.1.8
xxhash==3.5.0
multiprocess==0.70.16
sentry-sdk==2.28.0
python-dateutil==2.9.0.post0
pyarrow==20.0.0
annotated-types==0.6.0
pydantic_core==2.27.1
pyasn1_modules==0.4.2
pyasn1==0.6.1
llm-cmd-comp==1.1.1
tqdm==4.67.1
pyparsing==3.2.3
wcwidth==0.2.13
llm==0.25
networkx==3.4.2
conda_package_streaming==0.11.0
zstandard==0.23.0
nvidia-nvtx-cu12==12.6.77
python-dotenv==1.1.0
pydantic_cli==9.1.0
tabulate==0.9.0
nvidia-cuda-nvrtc-cu12==12.6.77
frozendict==2.4.2
google-ai-generativelanguage==0.6.15
truststore==0.10.0
Brotli==1.0.9
Markdown==3.8
nvidia-cusparse-cu12==12.5.4.2
llm-groq==0.8
puremagic==1.29
tzlocal==5.3.1
mpmath==1.3.0
python-ulid==3.0.0
sqlite-migrate==0.1b0
jiter==0.9.0
llm-cmd==0.2a0
htmldate==1.9.3
condense-json==0.1.2
dill==0.3.8
atroposlib==0.2.1
math-verify==0.7.0
MarkupSafe==3.0.2
jsonlines==4.0.0
yarl==1.20.0
httptools==0.6.4
jusText==3.0.2
multidict==6.4.3
nvidia-cudnn-cu12==9.5.1.17
Farama-Notifications==0.0.4
aiohappyeyeballs==2.6.1
openai==1.78.0
ruamel.yaml==0.18.10
PyYAML==6.0.2
Pygments==2.19.1
PySocks==1.7.1
boltons==24.1.0
propcache==0.3.1
conda-package-handling==2.4.0
anyio==4.9.0
conda-content-trust==0.2.0
google-api-core==2.25.0rc0
wheel==0.45.1
fsspec==2025.3.0
libmambapy==2.0.5
smmap==5.0.2
markdown-it-py==2.2.0
gitdb==4.0.12
prompt_toolkit==3.0.51
regex==2024.11.6
nvidia-nvjitlink-cu12==12.6.85
proto-plus==1.26.1
google-generativeai==0.8.5
cffi==1.17.1
torch==2.7.0
certifi==2025.4.26
llm-jq==0.1.1
lxml==5.4.0
trafilatura==2.0.0
antlr4-python3-runtime==4.13.2
platformdirs==4.3.7
sqlite-utils==3.38
Jinja2==3.1.6
lxml_html_clean==0.4.2
typing_extensions==4.12.2
huggingface-hub==0.31.2
joblib==1.5.0
triton==3.3.0
rich==13.9.4
nvidia-curand-cu12==10.3.7.77
aiohttp==3.11.18
distro==1.9.0
pip==25.0
urllib3==2.3.0
setproctitle==1.3.6
pandas==2.2.3
nvidia-cufile-cu12==1.11.1.6
watchfiles==1.0.5
jsonpatch==1.33
ruamel.yaml.clib==0.2.12
conda-anaconda-telemetry==0.1.2
sniffio==1.3.1
nvidia-cuda-runtime-cu12==12.6.77
pycosat==0.6.6
conda-anaconda-tos==0.1.3
latex2sympy2_extended==1.10.1
grpcio-status==1.71.0
conda-libmamba-solver==25.4.0
transformers==4.51.3
nltk==3.9.1
sympy==1.14.0
gymnasium==1.1.1
groq==0.24.0
hf_transfer==0.1.9
courlan==1.3.2
google-auth==2.39.0
pytz==2025.2
docker-pycreds==0.4.0
filelock==3.18.0
dateparser==1.2.1
mdurl==0.1.0
GitPython==3.1.44
uritemplate==4.1.1
menuinst==2.2.0
pydantic==2.10.3
httpx==0.28.1
googleapis-common-protos==1.70.0
frozenlist==1.6.0
google-auth-httplib2==0.2.0
aiosignal==1.3.2
websockets==15.0.1
tld==0.13
rsa==4.9.1
httpcore==1.0.9
setuptools==78.1.1
tenacity==9.1.2
httplib2==0.22.0
sqlite-fts4==1.0.3
cryptography==44.0.1
fastapi==0.115.12
archspec==0.2.3
pluggy==1.5.0
idna==3.7
tzdata==2025.2
tokenizers==0.21.1
nvidia-cusparselt-cu12==0.6.3
charset-normalizer==3.4.2
safetensors==0.5.3

View file

@ -0,0 +1,44 @@
{
"os": "Linux-5.15.167.4-microsoft-standard-WSL2-x86_64-with-glibc2.40",
"python": "CPython 3.13.2",
"startedAt": "2025-05-19T00:45:35.921074Z",
"args": [
"process",
"--openai.base_url",
"https://api.x.ai/v1",
"--openai.api_key",
"xai-cjDD6fd8ArfNJgc5Xnz3AjiJNuUxMuaD1rcpTAOR4RRh0SSO8v2KBwjYwUuEI2kFQxQOkUx943keKiUF",
"--openai.model_name",
"grok-3-beta",
"--env.group_size",
"2",
"--env.data_path",
"humor_dataset.jsonl"
],
"program": "/home/kirill/hackathons/atropos-nrs15h/environments/hack0/llm_humor_server.py",
"codePath": "environments/hack0/llm_humor_server.py",
"git": {
"remote": "git@github.com:kirilligum/atropos-nrs15h.git",
"commit": "db1e68d2abd2f64515325da6aac5a0484efd1741"
},
"email": "kirill.igum@gmail.com",
"root": "/home/kirill/hackathons/atropos-nrs15h/environments/hack0",
"host": "koldun",
"executable": "/home/kirill/miniconda3/bin/python",
"codePathLocal": "llm_humor_server.py",
"cpu_count": 7,
"cpu_count_logical": 14,
"disk": {
"/": {
"total": "1081101176832",
"used": "161502183424"
}
},
"memory": {
"total": "16480206848"
},
"cpu": {
"count": 7,
"countLogical": 14
}
}

View file

@ -0,0 +1 @@
{"train/completion_lengths_std":106.5,"train/completion_lengths_p95":1,"train/rollouts":{"sha256":"ddb3d4b414e5f8352fd4e28756535b594208247e70ca5ff50b11d18480ea0a65","size":8804,"artifact_path":"wandb-client-artifact://23v2xzjnymg72h1trblaybspr7kvzrvhm097wbmytwgbidf4qxl3vpvdwqi5urbrttzsqthwjqi28lc5awvcqca29jgnxecymti1kthxlfmiu84r3f0k62lmzczxpn8h/train/rollouts.table.json","_latest_artifact_path":"wandb-client-artifact://snf0xzi8g2ek14j1i8j4ydft6t1bqnxnorv0cx37x13v5kdyfwxco4l3t29re0ft5uew90x6subpcrw9xx35s8pbv3afww9muy62lmicxxgxmm5qopfe3kc71sjxfwwr:latest/train/rollouts.table.json","path":"media/table/train/rollouts_0_ddb3d4b414e5f8352fd4.table.json","ncols":2,"nrows":1,"_type":"table-file"},"server/server_0_request_time_std":2.75839364528656,"server/server_0_average_num_attempts":1,"server/server_0_request_time_avg":43.25876533985138,"_runtime":88.049322698,"train/completion_lengths":2036.5,"server/server_0_request_time_99p":45.961991112232205,"_timestamp":1.747615623316455e+09,"train/completion_lengths_max":2143,"train/completion_lengths_min":1930,"_step":1,"_wandb":{"runtime":88}}

View file

@ -0,0 +1 @@
/home/kirill/.cache/wandb/logs/core-debug-20250518_174535.log

View file

@ -0,0 +1,15 @@
{"time":"2025-05-18T17:45:35.922743021-07:00","level":"INFO","msg":"stream: starting","core version":"0.19.11","symlink path":"/home/kirill/hackathons/atropos-nrs15h/environments/hack0/wandb/run-20250518_174535-c91dmxez/logs/debug-core.log"}
{"time":"2025-05-18T17:45:36.140593855-07:00","level":"INFO","msg":"created new stream","id":"c91dmxez"}
{"time":"2025-05-18T17:45:36.140664557-07:00","level":"INFO","msg":"stream: started","id":"c91dmxez"}
{"time":"2025-05-18T17:45:36.140808847-07:00","level":"INFO","msg":"handler: started","stream_id":"c91dmxez"}
{"time":"2025-05-18T17:45:36.140842138-07:00","level":"INFO","msg":"sender: started","stream_id":"c91dmxez"}
{"time":"2025-05-18T17:45:36.140855771-07:00","level":"INFO","msg":"writer: Do: started","stream_id":"c91dmxez"}
{"time":"2025-05-18T17:45:36.259558417-07:00","level":"INFO","msg":"Starting system monitor"}
{"time":"2025-05-18T17:47:03.970372545-07:00","level":"INFO","msg":"stream: closing","id":"c91dmxez"}
{"time":"2025-05-18T17:47:03.970434503-07:00","level":"INFO","msg":"Stopping system monitor"}
{"time":"2025-05-18T17:47:03.970514355-07:00","level":"INFO","msg":"Stopped system monitor"}
{"time":"2025-05-18T17:47:04.391607503-07:00","level":"INFO","msg":"fileTransfer: Close: file transfer manager closed"}
{"time":"2025-05-18T17:47:04.535988574-07:00","level":"INFO","msg":"handler: closed","stream_id":"c91dmxez"}
{"time":"2025-05-18T17:47:04.536032472-07:00","level":"INFO","msg":"writer: Close: closed","stream_id":"c91dmxez"}
{"time":"2025-05-18T17:47:04.536062076-07:00","level":"INFO","msg":"sender: closed","stream_id":"c91dmxez"}
{"time":"2025-05-18T17:47:04.536084978-07:00","level":"INFO","msg":"stream: closed","id":"c91dmxez"}

View file

@ -0,0 +1,23 @@
2025-05-18 17:45:35,917 INFO MainThread:68123 [wandb_setup.py:_flush():70] Current SDK version is 0.19.11
2025-05-18 17:45:35,917 INFO MainThread:68123 [wandb_setup.py:_flush():70] Configure stats pid to 68123
2025-05-18 17:45:35,918 INFO MainThread:68123 [wandb_setup.py:_flush():70] Loading settings from /home/kirill/.config/wandb/settings
2025-05-18 17:45:35,918 INFO MainThread:68123 [wandb_setup.py:_flush():70] Loading settings from /home/kirill/hackathons/atropos-nrs15h/environments/hack0/wandb/settings
2025-05-18 17:45:35,918 INFO MainThread:68123 [wandb_setup.py:_flush():70] Loading settings from environment variables
2025-05-18 17:45:35,918 INFO MainThread:68123 [wandb_init.py:setup_run_log_directory():724] Logging user logs to /home/kirill/hackathons/atropos-nrs15h/environments/hack0/wandb/run-20250518_174535-c91dmxez/logs/debug.log
2025-05-18 17:45:35,918 INFO MainThread:68123 [wandb_init.py:setup_run_log_directory():725] Logging internal logs to /home/kirill/hackathons/atropos-nrs15h/environments/hack0/wandb/run-20250518_174535-c91dmxez/logs/debug-internal.log
2025-05-18 17:45:35,918 INFO MainThread:68123 [wandb_init.py:init():852] calling init triggers
2025-05-18 17:45:35,918 INFO MainThread:68123 [wandb_init.py:init():857] wandb.init called with sweep_config: {}
config: {'group_size': 2, 'max_num_workers': -1, 'max_eval_workers': 16, 'max_num_workers_per_node': 8, 'steps_per_eval': 100, 'max_token_length': 2048, 'eval_handling': <EvalHandlingEnum.STOP_TRAIN: 'STOP_TRAIN'>, 'eval_limit_ratio': 0.5, 'inference_weight': 1.0, 'batch_size': -1, 'max_batches_offpolicy': 3, 'tokenizer_name': 'NousResearch/DeepHermes-3-Llama-3-3B-Preview', 'use_wandb': True, 'rollout_server_url': 'http://localhost:8000', 'total_steps': 2, 'wandb_name': None, 'num_rollouts_to_keep': 32, 'num_rollouts_per_group_for_logging': 1, 'ensure_scores_are_not_same': False, 'data_path_to_save_groups': 'data/humor_2.jsonl', 'min_items_sent_before_logging': 2, 'include_messages': True, 'data_path': 'humor_dataset.jsonl', '_wandb': {}}
2025-05-18 17:45:35,918 INFO MainThread:68123 [wandb_init.py:init():893] starting backend
2025-05-18 17:45:35,918 INFO MainThread:68123 [wandb_init.py:init():897] sending inform_init request
2025-05-18 17:45:35,920 INFO MainThread:68123 [backend.py:_multiprocessing_setup():101] multiprocessing start_methods=fork,spawn,forkserver, using: spawn
2025-05-18 17:45:35,920 INFO MainThread:68123 [wandb_init.py:init():907] backend started and connected
2025-05-18 17:45:35,921 INFO MainThread:68123 [wandb_init.py:init():1005] updated telemetry
2025-05-18 17:45:35,925 INFO MainThread:68123 [wandb_init.py:init():1029] communicating run to backend with 90.0 second timeout
2025-05-18 17:45:36,257 INFO MainThread:68123 [wandb_init.py:init():1104] starting run threads in backend
2025-05-18 17:45:36,322 INFO MainThread:68123 [wandb_run.py:_console_start():2573] atexit reg
2025-05-18 17:45:36,322 INFO MainThread:68123 [wandb_run.py:_redirect():2421] redirect: wrap_raw
2025-05-18 17:45:36,322 INFO MainThread:68123 [wandb_run.py:_redirect():2490] Wrapping output streams.
2025-05-18 17:45:36,322 INFO MainThread:68123 [wandb_run.py:_redirect():2513] Redirects installed.
2025-05-18 17:45:36,323 INFO MainThread:68123 [wandb_init.py:init():1150] run started, returning control to user process
2025-05-18 17:47:03,969 INFO MsgRouterThr:68123 [mailbox.py:close():129] [no run ID] Closing mailbox, abandoning 2 handles.

View file

@ -0,0 +1,79 @@
_wandb:
value:
cli_version: 0.19.11
m: []
python_version: 3.13.2
t:
"1":
- 1
- 11
- 49
- 51
- 55
- 95
"2":
- 1
- 11
- 49
- 51
- 55
- 95
"3":
- 13
- 16
- 23
- 55
- 61
"4": 3.13.2
"5": 0.19.11
"6": 4.51.3
"8":
- 5
"12": 0.19.11
"13": linux-x86_64
batch_size:
value: -1
data_path:
value: humor_dataset.jsonl
data_path_to_save_groups:
value: data/humor_3.jsonl
ensure_scores_are_not_same:
value: false
eval_handling:
value: STOP_TRAIN
eval_limit_ratio:
value: 0.5
group_size:
value: 2
include_messages:
value: true
inference_weight:
value: 1
max_batches_offpolicy:
value: 3
max_eval_workers:
value: 16
max_num_workers:
value: -1
max_num_workers_per_node:
value: 8
max_token_length:
value: 2048
min_items_sent_before_logging:
value: 2
num_rollouts_per_group_for_logging:
value: 1
num_rollouts_to_keep:
value: 32
rollout_server_url:
value: http://localhost:8000
steps_per_eval:
value: 100
tokenizer_name:
value: NousResearch/DeepHermes-3-Llama-3-3B-Preview
total_steps:
value: 2
use_wandb:
value: true
wandb_name:
value: null

View file

@ -0,0 +1,9 @@
Starting to process 2 groups...
Processing group 1/2
Wrote scored group to data/humor_3.jsonl
Successfully processed group 1/2
Processing group 2/2
Wrote scored group to data/humor_3.jsonl
Successfully processed group 2/2
Completed processing 2 groups
Successfully generated HTML file: /home/kirill/hackathons/atropos-nrs15h/environments/hack0/data/humor_3.html

View file

@ -0,0 +1,178 @@
google-api-python-client==2.169.0
aiofiles==24.1.0
datasets==3.6.0
requests==2.32.3
nvidia-cusolver-cu12==11.7.1.2
babel==2.17.0
grpcio==1.71.0
polars==1.29.0
packaging==24.2
anaconda-anon-usage==0.7.0
nvidia-cufft-cu12==11.3.0.4
wandb==0.19.11
attrs==25.3.0
starlette==0.46.2
six==1.17.0
cloudpickle==3.1.1
numpy==2.2.6
pycparser==2.21
protobuf==5.29.4
cachetools==5.5.2
nvidia-cublas-cu12==12.6.4.1
uvloop==0.21.0
jsonpointer==2.1
psutil==7.0.0
conda==25.3.1
nvidia-nccl-cu12==2.26.2
h11==0.16.0
click-default-group==1.2.4
nvidia-cuda-cupti-cu12==12.6.80
uvicorn==0.34.2
click==8.1.8
xxhash==3.5.0
multiprocess==0.70.16
sentry-sdk==2.28.0
python-dateutil==2.9.0.post0
pyarrow==20.0.0
annotated-types==0.6.0
pydantic_core==2.27.1
pyasn1_modules==0.4.2
pyasn1==0.6.1
llm-cmd-comp==1.1.1
tqdm==4.67.1
pyparsing==3.2.3
wcwidth==0.2.13
llm==0.25
networkx==3.4.2
conda_package_streaming==0.11.0
zstandard==0.23.0
nvidia-nvtx-cu12==12.6.77
python-dotenv==1.1.0
pydantic_cli==9.1.0
tabulate==0.9.0
nvidia-cuda-nvrtc-cu12==12.6.77
frozendict==2.4.2
google-ai-generativelanguage==0.6.15
truststore==0.10.0
Brotli==1.0.9
Markdown==3.8
nvidia-cusparse-cu12==12.5.4.2
llm-groq==0.8
puremagic==1.29
tzlocal==5.3.1
mpmath==1.3.0
python-ulid==3.0.0
sqlite-migrate==0.1b0
jiter==0.9.0
llm-cmd==0.2a0
htmldate==1.9.3
condense-json==0.1.2
dill==0.3.8
atroposlib==0.2.1
math-verify==0.7.0
MarkupSafe==3.0.2
jsonlines==4.0.0
yarl==1.20.0
httptools==0.6.4
jusText==3.0.2
multidict==6.4.3
nvidia-cudnn-cu12==9.5.1.17
Farama-Notifications==0.0.4
aiohappyeyeballs==2.6.1
openai==1.78.0
ruamel.yaml==0.18.10
PyYAML==6.0.2
Pygments==2.19.1
PySocks==1.7.1
boltons==24.1.0
propcache==0.3.1
conda-package-handling==2.4.0
anyio==4.9.0
conda-content-trust==0.2.0
google-api-core==2.25.0rc0
wheel==0.45.1
fsspec==2025.3.0
libmambapy==2.0.5
smmap==5.0.2
markdown-it-py==2.2.0
gitdb==4.0.12
prompt_toolkit==3.0.51
regex==2024.11.6
nvidia-nvjitlink-cu12==12.6.85
proto-plus==1.26.1
google-generativeai==0.8.5
cffi==1.17.1
torch==2.7.0
certifi==2025.4.26
llm-jq==0.1.1
lxml==5.4.0
trafilatura==2.0.0
antlr4-python3-runtime==4.13.2
platformdirs==4.3.7
sqlite-utils==3.38
Jinja2==3.1.6
lxml_html_clean==0.4.2
typing_extensions==4.12.2
huggingface-hub==0.31.2
joblib==1.5.0
triton==3.3.0
rich==13.9.4
nvidia-curand-cu12==10.3.7.77
aiohttp==3.11.18
distro==1.9.0
pip==25.0
urllib3==2.3.0
setproctitle==1.3.6
pandas==2.2.3
nvidia-cufile-cu12==1.11.1.6
watchfiles==1.0.5
jsonpatch==1.33
ruamel.yaml.clib==0.2.12
conda-anaconda-telemetry==0.1.2
sniffio==1.3.1
nvidia-cuda-runtime-cu12==12.6.77
pycosat==0.6.6
conda-anaconda-tos==0.1.3
latex2sympy2_extended==1.10.1
grpcio-status==1.71.0
conda-libmamba-solver==25.4.0
transformers==4.51.3
nltk==3.9.1
sympy==1.14.0
gymnasium==1.1.1
groq==0.24.0
hf_transfer==0.1.9
courlan==1.3.2
google-auth==2.39.0
pytz==2025.2
docker-pycreds==0.4.0
filelock==3.18.0
dateparser==1.2.1
mdurl==0.1.0
GitPython==3.1.44
uritemplate==4.1.1
menuinst==2.2.0
pydantic==2.10.3
httpx==0.28.1
googleapis-common-protos==1.70.0
frozenlist==1.6.0
google-auth-httplib2==0.2.0
aiosignal==1.3.2
websockets==15.0.1
tld==0.13
rsa==4.9.1
httpcore==1.0.9
setuptools==78.1.1
tenacity==9.1.2
httplib2==0.22.0
sqlite-fts4==1.0.3
cryptography==44.0.1
fastapi==0.115.12
archspec==0.2.3
pluggy==1.5.0
idna==3.7
tzdata==2025.2
tokenizers==0.21.1
nvidia-cusparselt-cu12==0.6.3
charset-normalizer==3.4.2
safetensors==0.5.3

View file

@ -0,0 +1,44 @@
{
"os": "Linux-5.15.167.4-microsoft-standard-WSL2-x86_64-with-glibc2.40",
"python": "CPython 3.13.2",
"startedAt": "2025-05-19T00:48:35.065253Z",
"args": [
"process",
"--openai.base_url",
"https://api.x.ai/v1",
"--openai.api_key",
"xai-cjDD6fd8ArfNJgc5Xnz3AjiJNuUxMuaD1rcpTAOR4RRh0SSO8v2KBwjYwUuEI2kFQxQOkUx943keKiUF",
"--openai.model_name",
"grok-3-beta",
"--env.group_size",
"2",
"--env.data_path",
"humor_dataset.jsonl"
],
"program": "/home/kirill/hackathons/atropos-nrs15h/environments/hack0/llm_humor_server.py",
"codePath": "environments/hack0/llm_humor_server.py",
"git": {
"remote": "git@github.com:kirilligum/atropos-nrs15h.git",
"commit": "96043a968f0afc34b8c1fb9dd104056fc6113e8b"
},
"email": "kirill.igum@gmail.com",
"root": "/home/kirill/hackathons/atropos-nrs15h/environments/hack0",
"host": "koldun",
"executable": "/home/kirill/miniconda3/bin/python",
"codePathLocal": "llm_humor_server.py",
"cpu_count": 7,
"cpu_count_logical": 14,
"disk": {
"/": {
"total": "1081101176832",
"used": "161502572544"
}
},
"memory": {
"total": "16480206848"
},
"cpu": {
"count": 7,
"countLogical": 14
}
}

View file

@ -0,0 +1 @@
{"train/completion_lengths_min":2102,"server/server_0_request_time_std":21.22465526590049,"train/completion_lengths":2144,"train/completion_lengths_max":2186,"server/server_0_request_time_99p":56.2853569149971,"server/server_0_request_time_avg":18.77630599339803,"server/server_0_average_num_attempts":1,"_step":1,"train/completion_lengths_p95":1,"_runtime":114.552826332,"train/rollouts":{"_type":"table-file","sha256":"9ae052d971ad0b81fb1e59ce189f395f673d5cf00a63b6a65621f1a19ec8f215","size":9249,"artifact_path":"wandb-client-artifact://l6ftszpxbw1dpa02ae8h0vsn6o0xdzp4t0aktnzyzwzijfx65v413rfedky7slp0rp0bam6f299b6hykeaz6t3b3tg10uhmo8tes4weidl9kbg0pyl5w7zpx6ivsdcfv/train/rollouts.table.json","_latest_artifact_path":"wandb-client-artifact://9vkgz8vzqrbwyz107iy6qmuequd5twgvutxjbeyihqcjuphsom39abn3fmnqhxqz9c9z0igp2xlwtwmyoorjnuxg3v06tnvw44nk2yb9j7xr0v73krldn5vwqjht2r5o:latest/train/rollouts.table.json","path":"media/table/train/rollouts_0_9ae052d971ad0b81fb1e.table.json","ncols":2,"nrows":1},"train/completion_lengths_std":42,"_wandb":{"runtime":114},"_timestamp":1.7476158286909504e+09}

View file

@ -0,0 +1 @@
/home/kirill/.cache/wandb/logs/core-debug-20250518_174834.log

View file

@ -0,0 +1,15 @@
{"time":"2025-05-18T17:48:35.067285373-07:00","level":"INFO","msg":"stream: starting","core version":"0.19.11","symlink path":"/home/kirill/hackathons/atropos-nrs15h/environments/hack0/wandb/run-20250518_174835-2wfxhi28/logs/debug-core.log"}
{"time":"2025-05-18T17:48:35.281397518-07:00","level":"INFO","msg":"created new stream","id":"2wfxhi28"}
{"time":"2025-05-18T17:48:35.281429663-07:00","level":"INFO","msg":"stream: started","id":"2wfxhi28"}
{"time":"2025-05-18T17:48:35.281465438-07:00","level":"INFO","msg":"writer: Do: started","stream_id":"2wfxhi28"}
{"time":"2025-05-18T17:48:35.281624943-07:00","level":"INFO","msg":"handler: started","stream_id":"2wfxhi28"}
{"time":"2025-05-18T17:48:35.28163269-07:00","level":"INFO","msg":"sender: started","stream_id":"2wfxhi28"}
{"time":"2025-05-18T17:48:35.435106526-07:00","level":"INFO","msg":"Starting system monitor"}
{"time":"2025-05-18T17:50:29.617997906-07:00","level":"INFO","msg":"stream: closing","id":"2wfxhi28"}
{"time":"2025-05-18T17:50:29.618126207-07:00","level":"INFO","msg":"Stopping system monitor"}
{"time":"2025-05-18T17:50:29.61816387-07:00","level":"INFO","msg":"Stopped system monitor"}
{"time":"2025-05-18T17:50:30.028808332-07:00","level":"INFO","msg":"fileTransfer: Close: file transfer manager closed"}
{"time":"2025-05-18T17:50:30.174820159-07:00","level":"INFO","msg":"handler: closed","stream_id":"2wfxhi28"}
{"time":"2025-05-18T17:50:30.174870994-07:00","level":"INFO","msg":"sender: closed","stream_id":"2wfxhi28"}
{"time":"2025-05-18T17:50:30.17484606-07:00","level":"INFO","msg":"writer: Close: closed","stream_id":"2wfxhi28"}
{"time":"2025-05-18T17:50:30.175032671-07:00","level":"INFO","msg":"stream: closed","id":"2wfxhi28"}

View file

@ -0,0 +1,23 @@
2025-05-18 17:48:35,060 INFO MainThread:68462 [wandb_setup.py:_flush():70] Current SDK version is 0.19.11
2025-05-18 17:48:35,060 INFO MainThread:68462 [wandb_setup.py:_flush():70] Configure stats pid to 68462
2025-05-18 17:48:35,060 INFO MainThread:68462 [wandb_setup.py:_flush():70] Loading settings from /home/kirill/.config/wandb/settings
2025-05-18 17:48:35,060 INFO MainThread:68462 [wandb_setup.py:_flush():70] Loading settings from /home/kirill/hackathons/atropos-nrs15h/environments/hack0/wandb/settings
2025-05-18 17:48:35,060 INFO MainThread:68462 [wandb_setup.py:_flush():70] Loading settings from environment variables
2025-05-18 17:48:35,060 INFO MainThread:68462 [wandb_init.py:setup_run_log_directory():724] Logging user logs to /home/kirill/hackathons/atropos-nrs15h/environments/hack0/wandb/run-20250518_174835-2wfxhi28/logs/debug.log
2025-05-18 17:48:35,060 INFO MainThread:68462 [wandb_init.py:setup_run_log_directory():725] Logging internal logs to /home/kirill/hackathons/atropos-nrs15h/environments/hack0/wandb/run-20250518_174835-2wfxhi28/logs/debug-internal.log
2025-05-18 17:48:35,060 INFO MainThread:68462 [wandb_init.py:init():852] calling init triggers
2025-05-18 17:48:35,060 INFO MainThread:68462 [wandb_init.py:init():857] wandb.init called with sweep_config: {}
config: {'group_size': 2, 'max_num_workers': -1, 'max_eval_workers': 16, 'max_num_workers_per_node': 8, 'steps_per_eval': 100, 'max_token_length': 2048, 'eval_handling': <EvalHandlingEnum.STOP_TRAIN: 'STOP_TRAIN'>, 'eval_limit_ratio': 0.5, 'inference_weight': 1.0, 'batch_size': -1, 'max_batches_offpolicy': 3, 'tokenizer_name': 'NousResearch/DeepHermes-3-Llama-3-3B-Preview', 'use_wandb': True, 'rollout_server_url': 'http://localhost:8000', 'total_steps': 2, 'wandb_name': None, 'num_rollouts_to_keep': 32, 'num_rollouts_per_group_for_logging': 1, 'ensure_scores_are_not_same': False, 'data_path_to_save_groups': 'data/humor_3.jsonl', 'min_items_sent_before_logging': 2, 'include_messages': True, 'data_path': 'humor_dataset.jsonl', '_wandb': {}}
2025-05-18 17:48:35,060 INFO MainThread:68462 [wandb_init.py:init():893] starting backend
2025-05-18 17:48:35,060 INFO MainThread:68462 [wandb_init.py:init():897] sending inform_init request
2025-05-18 17:48:35,064 INFO MainThread:68462 [backend.py:_multiprocessing_setup():101] multiprocessing start_methods=fork,spawn,forkserver, using: spawn
2025-05-18 17:48:35,065 INFO MainThread:68462 [wandb_init.py:init():907] backend started and connected
2025-05-18 17:48:35,066 INFO MainThread:68462 [wandb_init.py:init():1005] updated telemetry
2025-05-18 17:48:35,070 INFO MainThread:68462 [wandb_init.py:init():1029] communicating run to backend with 90.0 second timeout
2025-05-18 17:48:35,433 INFO MainThread:68462 [wandb_init.py:init():1104] starting run threads in backend
2025-05-18 17:48:35,491 INFO MainThread:68462 [wandb_run.py:_console_start():2573] atexit reg
2025-05-18 17:48:35,491 INFO MainThread:68462 [wandb_run.py:_redirect():2421] redirect: wrap_raw
2025-05-18 17:48:35,491 INFO MainThread:68462 [wandb_run.py:_redirect():2490] Wrapping output streams.
2025-05-18 17:48:35,491 INFO MainThread:68462 [wandb_run.py:_redirect():2513] Redirects installed.
2025-05-18 17:48:35,492 INFO MainThread:68462 [wandb_init.py:init():1150] run started, returning control to user process
2025-05-18 17:50:29,616 INFO MsgRouterThr:68462 [mailbox.py:close():129] [no run ID] Closing mailbox, abandoning 2 handles.