davanstrien/rewire-poc / cc-4b /logs /executor.json
davanstrien's picture
download
raw
4.36 kB
{
"pipeline": [
{
"limit": -1,
"skip": 0,
"text_key": "text",
"id_key": "id",
"adapter": "<bound method BaseReader._default_adapter of \ud83d\udcd6 - READER: \ud83e\udd17 HuggingFace>",
"_empty_warning": false,
"default_metadata": null,
"dataset": "davanstrien/rewire-poc-sources-cc",
"dataset_options": {
"split": "train[0:400]"
},
"batch_size": 1000,
"doc_progress": false,
"streaming": false,
"shuffle_files": false,
"load_from_disk": false
},
{
"config": {
"server_type": "sglang",
"model_name_or_path": "Qwen/Qwen3.5-4B",
"model_max_context": 16384,
"use_chat": true,
"endpoint_url": null,
"api_key": null,
"metric_interval": 120,
"tp": 1,
"dp": 1,
"pp": 1,
"default_generation_params": {},
"rollouts_per_document": 1,
"max_concurrent_generations": 64,
"max_concurrent_documents": 64,
"request_timeout": null,
"model_kwargs": null,
"server_log_folder": null,
"master_port": 9810,
"http_client": "raw",
"server_start_timeout_sec": 60.0,
"server_ready_max_attempts": 300,
"server_ready_delay_sec": 5.0,
"server_start_max_retries": 1
},
"rollout_fn": "<function make_rollout.<locals>.rollout at 0x10bab6de0>",
"shared_context": null,
"metadata_key": "rollout_results",
"skip_bad_requests": false,
"output_writer": {
"compression": "snappy",
"output_folder": "DataFolder(path='buckets/davanstrien/rewire-poc/cc-4b/gen', fs=<huggingface_hub.hf_file_system.HfFileSystem object at 0x10b86e900>)",
"max_file_size": 8388608,
"file_id_counter": {},
"output_filename": "<string.Template object at 0x10ba10c50>",
"output_mg": "<datatrove.io.OutputFileManager object at 0x10b980380>",
"adapter": "<bound method DiskWriter._default_adapter of \ud83d\udcbd - WRITER: \ud83d\udcd2 Parquet>",
"expand_metadata": false,
"save_media_bytes": false,
"_writers": {},
"_batches": {},
"_file_counter": {},
"batch_size": 50,
"schema": null,
"use_content_defined_chunking": {
"min_chunk_size": 262144,
"max_chunk_size": 1048576,
"norm_level": 0
},
"write_page_index": true
},
"request_cache": "<datatrove.pipeline.inference.checkpointing.RequestCache object at 0x105c844d0>",
"checkpoint_manager": "<datatrove.pipeline.inference.checkpointing.CheckpointManager object at 0x10b5c4b30>",
"metrics": "Metric Name Lifetime (/sec) Recently (/sec)\n------------------------------------------------------------------------",
"queue_sizes": "No queue sizes tracked yet."
}
],
"logging_dir": "DataFolder(path='buckets/davanstrien/rewire-poc/cc-4b/logs', fs=<huggingface_hub.hf_file_system.HfFileSystem object at 0x10b86e900>)",
"skip_completed": true,
"randomize_start_duration": 0,
"tasks": 1,
"workers": 1,
"flavor": "a100-large",
"image": "lmsysorg/sglang:nightly-dev-cu13-20260709-074bb928",
"dependencies": [
"datatrove[io,processing] @ https://github.com/davanstrien/datatrove/archive/f7df218.tar.gz",
"sglang",
"datasets",
"orjson",
"aiosqlite",
"aiofiles"
],
"timeout": "1h",
"tasks_per_job": 1,
"depends": null,
"python": "3.12",
"namespace": null,
"labels": null,
"token": null,
"secrets": null,
"max_retries": 0,
"poll_interval": 15,
"run_on_dependency_fail": false,
"job_name": "rewire-cc-4b",
"_launched": false,
"world_size": 1
}

Xet Storage Details

Size:
4.36 kB
·
Xet hash:
ce5fcc7fdf042f1df33094c6a8794a61d7b7b67981b615f2bdb7caef47d5e63d

Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.