CatalogWantedMirrorsLedgerOperating StatementDocsStudio statusCard system
alt.freeweights.text / IndustryCorpus2_DataRater
From: BAAI <huggingface.co/BAAI>Newsgroups: alt.freeweights.textSubject:

IndustryCorpus2_DataRater

Date: 2026-11-15Commit: 298dcc5Signed: 2DC28CC7B26A283B ✓

Model with 568M parameters. Safetensors weights, 2.29 GB in all. Published by BAAI under Apache-2.0.

568M params2.29GBIndustryCorpus2_DataRatersafetensorspermissive
FactsParams568MSize2.29 GBFormatssafetensorsTasknot taggedLicenseApache-2.0From sourcepermissive
$ fw get BAAI/IndustryCorpus2_DataRater
# manifest sha256 efd15b94c955… · 2.29 GB · 13 files
$ fw verify ./IndustryCorpus2_DataRater BAAI/IndustryCorpus2_DataRater
From sourceHosted (signed)BAAI/IndustryCorpus2_DataRaterLast check: 2026-11-15 04:15 UTC
Run at homeGet help

Get help: custom fine-tune or private deploy. We review your setup before we quote.

Fine-tune of bge-m3.

Works with

Run8Fine-tune7Convert8Scan2

Runs with

Ranked by fit to this model (safetensors), then by GitHub stars.

TransformersLocalAItinygradTriton Inference ServerBentoMLCTranslate2ExLlamaV2LitServe
OverviewModel cardFilesUse itLicenseProvenanceModel tree

Overview

Size
2.29 GB
Weights
2.27 GB (safetensors)
Parameters
568M
Formats
safetensors
Task
not tagged
License
Apache-2.0 (permissive)
Files
13, each with a sha256
Publisher
BAAI
Added
2026-11-15
Section
alt.freeweights.text
Lineage

Original release from the publisher.

Base model: BAAI/bge-m3

Where these bytes come from
Publisher at the pinned commit→You

Sources are tried in order. A file counts only when its size and sha256 match the manifest. Sources are checked when you ask for them, not on a timer.

Sources
BadgeHostLast checkResult
From sourcehuggingface.co2026-11-15 04:15 UTCpassed

Model card

From the publisher at commit 298dcc5. Shown as plain text with links; images open at the publisher.

本模型为数据集BAAI/IndustryCorpus2的质量评估模型，用于从语义一致性，信息密度，教育属性等维度评估预训练数据的质量。

按照我们的定义并经过实验，3分以上是相对高质量数据，4分以上绝对高质量数据，可以根据数据量按需所取。

为什么要筛选低质量的数据#

下面是从数据中抽取的低质量数据，可以看到这种数据对模型的学习是有害无益的

  {"text": "\\_\\__\n\nTranslated from *Chinese Journal of Biochemistry and Molecular Biology*, 2007, 23(2): 154--159 \\[译自:中国生物化学与分子生物学报\\]\n"}
  
  {"text": "#ifndef _IMGBMP_H_\n#define _IMGBMP_H_\n\n#ifdef __cplusplus\nextern \"C\" {\n#endif\n\nconst uint8_t bmp[]={\n\\/\\/--  调入了一幅图像：D:\\我的文档\\My Pictures\\12864-555.bmp  --*\\/\n\\/\\/--  宽度x高度=128x64  --\n0x00,0x06,0x0A,0xFE,0x0A,0xC6,0x00,0xE0,0x00,0xF0,0x00,0xF8,0x00,0x00,0x00,0x00,\n0x00,0x00,0xFE,0x7D,0xBB,0xC7,0xEF,0xEF,0xEF,0xEF,0xEF,0xEF,0xEF,0xC7,0xBB,0x7D,\n0xFE,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,\n0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,\n0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,\n0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x08,\n0x0C,0xFE,0xFE,0x0C,0x08,0x20,0x60,0xFE,0xFE,0x60,0x20,0x00,0x00,0x00,0x78,0x48,\n0xFE,0x82,0xBA,0xBA,0x82,0xBA,0xBA,0x82,0xBA,0xBA,0x82,0xBA,0xBA,0x82,0xFE,0x00,\n0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,\n0x00,0x00,0x00,0x01,0x01,0x01,0x01,0x01,0x01,0x01,0x01,0x01,0x01,0x01,0x01,0x01,\n0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,\n0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,\n0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,\n0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,\n0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,\n0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,\n0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,\n0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0xFE,0xFF,\n0x03,0x03,0x03,0x03,0x03,0x03,0x03,0x03,0x03,0xFF,0xFF,0x00,0x00,0xFE,0xFF,0x03,\n0x03,0x03,0x03,0x03,0x03,0x03,0x03,0x03,0xFF,0xFE,0x00,0x00,0x00,0x00,0xC0,0xC0,\n0xC0,0x00,0x00,0x00,0x00,0xFE,0xFF,0x03,0x03,0x03,0x03,0x03,0x03,0x03,0x03,0x03,\n0xFF,0xFE,0x00,0x00,0xFE,0xFF,0x03,0x03,0x03,0x03,0x03,0x03,0x03,0x03,0x03,0xFF,\n0xFE,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,\n0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,\n0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,\n0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0xFF,0xFF,\n0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0xFF,0xFF,0x00,0x00,0xFF,0xFF,0x0C,\n0x0C,0x0C,0x0C,0x0C,0x0C,0x0C,0x0C,0x0C,0xFF,0xFF,0x00,0x00,0x00,0x00,0xE1,0xE1,\n0xE1,0x00,0x00,0x00,0x00,0xFF,0xFF,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,\n0xFF,0xFF,0x00,0x00,0xFF,0xFF,0x0C,0x0C,0x0C,0x0C,0x0C,0x0C,0x0C,0x0C,0x0C,0xFF,\n0xFF,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,\n0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,\n0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,\n0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x0F,0x1F,\n0x18,0x18,0x18,0x18,0x18,0x18,0x18,0x18,0x18,0x1F,0x0F,0x00,0x00,0x0F,0x1F,0x18,\n0x18,0x18,0x18,0x18,0x18,0x18,0x18,0x18,0x1F,0x0F,0x00,0x00,0x00,0x00,0x00,0x00,\n0x00,0x00,0x00,0x00,0x00,0x0F,0x1F,0x18,0x18,0x18,0x18,0x18,0x18,0x18,0x18,0x18,\n0x1F,0x0F,0x00,0x00,0x0F,0x1F,0x18,0x18,0x18,0x18,0x18,0x18,0x18,0x18,0x18,0x1F,\n0x0F,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,\n0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,\n0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,\n0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,\n0x00,0x00,0x00,0x00,0x00,0xE2,0x92,0x8A,0x86,0x00,0x00,0x7C,0x82,0x82,0x82,0x7C,\n0x00,0xFE,0x00,0x82,0x92,0xAA,0xC6,0x00,0x00,0xC0,0xC0,0x00,0x7C,0x82,0x82,0x82,\n0x7C,0x00,0x00,0x02,0x02,0x02,0xFE,0x00,0x00,0xC0,0xC0,0x00,0x7C,0x82,0x82,0x82,\n0x7C,0x00,0x00,0xFE,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,\n0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,\n0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,\n0x00,0x24,0xA4,0x2E,0x24,0xE4,0x24,0x2E,0xA4,0x24,0x00,0x00,0x00,0xF8,0x4A,0x4C,\n0x48,0xF8,0x48,0x4C,0x4A,0xF8,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,\n0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,\n0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,\n0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,\n0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,\n0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0xC0,0x20,0x10,0x10,\n0x10,0x10,0x20,0xC0,0x00,0x00,0xC0,0x20,0x10,0x10,0x10,0x10,0x20,0xC0,0x00,0x00,\n0x00,0x12,0x0A,0x07,0x02,0x7F,0x02,0x07,0x0A,0x12,0x00,0x00,0x00,0x0B,0x0A,0x0A,\n0x0A,0x7F,0x0A,0x0A,0x0A,0x0B,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,\n0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,\n0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,\n0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,\n0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,\n0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x1F,0x20,0x40,0x40,\n0x40,0x50,0x20,0x5F,0x80,0x00,0x1F,0x20,0x40,0x40,0x40,0x50,0x20,0x5F,0x80,0x00,\n};  \n\n\n#ifdef __cplusplus\n}\n#endif\n\n#endif \\/\\/ _IMGBMP_H_ _SSD1306_16BIT_H_\n"}

数据构建#

数据来源：随机采样预训练语料

标签构建：设计数据打分细则，借助LLM模型进行多轮打分，筛选多轮打分分差小于2的数据

数据规模：20k打分数据，中英文比例1:1

数据prompt

  quality_prompt = """Below is an extract from a web page. Evaluate whether the page has a high natural language value and could be useful in an naturanl language task to train a good language model using the additive 5-point scoring system described below. Points are accumulated based on the satisfaction of each criterion:
  
  - Zero score if the content contains only some meaningless content or private content, such as some random code, http url or copyright information, personally identifiable information, binary encoding of images.
  - Add 1 point if the extract provides some basic information, even if it includes some useless contents like advertisements and promotional material.
  - Add another point if the extract is written in good style, semantically fluent, and free of repetitive content and grammatical errors.
  - Award a third point tf the extract has relatively complete semantic content, and is written in a good and fluent style, the entire content expresses something related to the same topic, rather than a patchwork of several unrelated items.
  - A fourth point is awarded if the extract has obvious educational or literary value, or provides a meaningful point or content, contributes to the learning of the topic, and is written in a clear and consistent style. It may be similar to a chapter in a textbook or tutorial, providing a lot of educational content, including exercises and solutions, with little to no superfluous information. The content is coherent and focused, which is valuable for structured learning.
  - A fifth point is awarded if the extract has outstanding educational value or is of very high information density, provides very high value and meaningful content, does not contain useless information, and is well suited for teaching or knowledge transfer. It contains detailed reasoning, has an easy-to-follow writing style, and can provide deep and thorough insights.
  
  
  The extract:
  <{EXAMPLE}>.
  
  After examining the extract:
  - Briefly justify your total score, up to 50 words.
  - Conclude with the score using the format: "Quality score:  <total points>"
  ...
  """

模型训练#

模型选型：与分类模型类似，我们同样使用的是0.5b规模的模型，并对比试验了beg-m3和qwen-0.5b，最终实验显示bge-m3综合表现最优

模型超参：基座bge-m3，全参数训练，lr=1e-5，batch_size=64, max_length = 2048

模型评估：在验证集上模型与GPT4对样本质量判定一致率为90%

Image: image-20240919142248242

  • 高质量数据带来的训练收益 我们为了验证高质量的数据是否能带来更高效的训练效率，在同一基座模型下，使用从未筛选之前的50b数据中抽取出高质量数据，可以认为两个数据的分布大体是一致的，进行自回归训练. 曲线中可以看到，经过高质量数据训练的模型14B的tokens可以达到普通数据50B的模型表现，高质量的数据可以极大的提升训练效率。 Image: image-20240919142732476

此外，高质量的数据可以作为预训练的退火阶段的数据加入到模型中，进一步拉升模型效果，为了验证这个猜测，我们在训练行业模型时候，在模型的退火阶段加入了筛选之后高质量数据和部分指令数据转成的预训练数据，可以看到极大提高了模型的表现。

Image: cpt_two_stage

最后，高质量的预训练语料中包含着丰富的高价值知识性内容，可以从中提取出指令数据进一步提升指令数据的丰富度和知识性，这也催发了IndustryInstruction项目的诞生，我们会在那里进行详细的说明。

how to use#


from transformers import (
    AutoConfig,
    AutoModelForSequenceClassification,
    AutoTokenizer)
import torch

model_dir = "BAAI/IndustryCorpus2_DataRater"

model = AutoModelForSequenceClassification.from_pretrained(
      model_dir,
      trust_remote_code=False,
      ignore_mismatched_sizes=False,
)
model.cuda()
model.eval()
tokenizer = AutoTokenizer.from_pretrained(
    model_dir,
    use_fast=True,
    token=None,
    trust_remote_code=False,
)

config = AutoConfig.from_pretrained(
    model_dir,
    finetuning_task="text-classification",
)

sentence = "黄龙国际中心位于杭州市西湖区学院路77号，地处杭州黄龙商务区核心位置。项目以“新经济、新生态、新黄龙”的理念和“知识城市创新区”的项目定位，引领着杭州城市的产业升级和创新。\n黄龙国际中心主打“知识盒子”建筑设计，拥有时尚、新潮的建筑立面，聚集不同功能的独立建筑物，打造出包容开放又具有链接性的空间。项目在2018年4月一举斩获开发项目行业最高奖——广厦奖。\n项目整体分四期开发。一期G、H、I三幢楼宇由坤和自主开发建成，于2015年4月投入使用，于2015年5月获得了美国绿色建筑委员会颁发的LEED金级认证，并成功吸引蚂蚁金服、美图、有赞等国内互联网创新巨头率先入驻。\n2016年初，坤和引进万科作为该项目股权合作方通过双方优质资源共享，在产品品质、创新等方面得到全方位提升。\n二期A、B幢由美国KPF设计，并于2018年4月一举获得房地产开发项目行业最高奖——广厦奖。A、B幢写字楼于2018年10月投入使用，B幢与全球领先的创造者社区及空间和服务提供商——WeWork正式签约。商业K-lab于2018年12月28日正式开业。\n项目三期E、F幢已于2020年投入使用。四期C、D幢，计划于2021年底投入使用。\n项目总占地面积约8.7万方，总建筑面积约40万平方米，涵盖9栋国际5A写字楼及8万方K-lab商业新物种，以知识为纽带，打造成一个集商务、商业、教育、文化、娱乐、艺术、餐饮、会展等于一体的完整城市性生态体系。项目全部投入使用后，年租金收入将超6亿元。\n"

result = tokenizer(
    [sentence],
    padding=False,
    max_length=2048,
    truncation=True,
    return_tensors="pt",
).to("cuda")
model_out = model(**result)
predict_score = model_out.logits.tolist()[0][0]
predict_score_rounded = round(predict_score)
# score=4.674278736114502

Citation#

If you use IndustryCorpus2_DataRater in your research, please cite the model:

@misc{shi2026industrycorpus2datarater,
  title={IndustryCorpus2_DataRater},
  author={Xiaofeng Shi and Hua Zhou and Lulu Zhao},
  year={2026},
  publisher={Hugging Face},
  doi={10.57967/hf/9611},
  url={https://huggingface.co/BAAI/IndustryCorpus2_DataRater}
}

Files

13 files · 2.29 GBWeights 2.27 GB (safetensors). Sizes and sha256 from the signed manifest.
Hide all 13 files with their sha256
FileSizesha256
.gitattributes1.57 KB34448b82c17d60fec9b65b1f093c115ddbaadc04beb1b0140b6bfed2e012a930
README.md13.4 KBe9f55098bbec0b582fbcd523bbf0958c85164c20006687991166b178dae02f6b
config.json860 Bba0390d80b7dc79bc8fab887b667e8cee61f4663e0c88f38d262f1a90dcbd427
img/.DS_Store6.15 KBa8a105b7950eb53ddbcb8f82b27d7445b6143f24cf239c8c4d19adf1c87dacbd
img/cpt_two_stage.png64.7 KB481ad8d7cf307b02e6af89e96ed5897a2c68fe82b151670d4115513952658cc7
img/quality-exp.png35.4 KB43e052dd07e3bb1b9de957bcd7ebf43c9aa6b0484eb986fea674554b592114c7
img/quality_train.png28.1 KBfa47313ff3906065c404fa704b2082e6d2c2ae0d807efc0ce84f9a71983d8f96
model.safetensors2.27 GB0425999578be322709ed81ed4d0b7083dcfdacc3b48ebbafeb89fe1f059f56d7
sentencepiece.bpe.model5.07 MBcfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
special_tokens_map.json964 B8c785abebea9ae3257b61681b4e6fd8365ceafde980c21970d001e834cf10835
tokenizer.json17.1 MB6710678b12670bc442b99edc952c4d996ae309a7020c1fa0096dd245c2faf790
tokenizer_config.json1.17 KB7e4c1cc848840aeccdd763458c18dd525eb0f795c992e00ebe9c28554e7db2d4
trainer_state.json34.7 KB15408e62378297583fe9587a3108dbf4cd29c7128ce5763fd63b0214a12bba59
VersionsEach time the pinned upstream commit or file list changed in this repo. Newest first.
Upstream commitDateContents
298dcc5d5a pinned now2026-11-15 04:15 UTC13 files · 2.29 GB

Use it

$ fw get BAAI/IndustryCorpus2_DataRater
# manifest sha256 efd15b94c955… · 2.29 GB · 13 files
$ fw verify ./IndustryCorpus2_DataRater BAAI/IndustryCorpus2_DataRater
fw.pyChecks the signature and every sha256, tries each source in order, resumes.
$ python3 fw.py get https://freeweights.allternit.com/manifests/BAAI/IndustryCorpus2_DataRater.json --dir ./IndustryCorpus2_DataRater $ python3 fw.py verify ./IndustryCorpus2_DataRater ./BAAI__IndustryCorpus2_DataRater.json
hfThe Hugging Face CLI with one setting. Same repo id.
$ HF_ENDPOINT=https://freeweights.allternit.com hf download BAAI/IndustryCorpus2_DataRater --revision 298dcc5d5ab4110363f1b06253c1a6f3e7f53bf8 --local-dir ./IndustryCorpus2_DataRater
Run at home

License

Name
Apache License 2.0
Id
Apache-2.0
Class
Permissive
Redistribution
yes
Listing
Hosted (signed)
Decided
License as declared by the publisher. No person read it. 2026-11-15
Text in
README.md
Text sha256
e9f55098bbec0b582fbcd523bbf0958c85164c20006687991166b178dae02f6b
Terms gate
No extra terms beyond the license.

Read the license text

Licenses are checked against the license table. A custom license passes only after a person reads it.

Provenance

From: BAAI <huggingface.co/BAAI>Newsgroups: alt.freeweights.textSubject: IndustryCorpus2_DataRaterDate: 2026-11-15Commit: 298dcc5Signed: 2DC28CC7B26A283B ✓
Upstream
BAAI/IndustryCorpus2_DataRater
Pinned commit
298dcc5d5ab4110363f1b06253c1a6f3e7f53bf8
Manifest
IndustryCorpus2_DataRater.json · .minisig
Manifest sha256
efd15b94c9554235aeb4685d3c544bd91c98b1f897679aa2c25ecd5ffe48c9b5
Signature
minisign, Ed25519, key 2DC28CC7B26A283B
Created
2026-11-15 04:15 UTC
Generator
freeweights-manifest/0.1.0
Takedowns
None.

How to verify

$ curl -O https://freeweights.allternit.com/manifests/BAAI/IndustryCorpus2_DataRater.json $ curl -O https://freeweights.allternit.com/manifests/BAAI/IndustryCorpus2_DataRater.json.minisig $ minisign -Vm IndustryCorpus2_DataRater.json -P RWQ7KGqyx4zCLXtUFOqvfv8L3g+8fpKJ1FsuxPyOYMF+4KufbUxKq1Oq $ shasum -a 256 <file> # compare with the Files tab

A manifest whose signature fails must be rejected. A file whose size or sha256 does not match must be discarded. Verify a download

Model tree

Parent

BAAI/bge-m3 · this model is a finetune of it

No model in the catalog names this one as its parent.

Get it
Download
$ HF_ENDPOINT=https://freeweights.allternit.com hf download BAAI/IndustryCorpus2_DataRater
Open in Freeweights Studio

Pull, verify and run it in the Allternit app on Desktop, web or phone.

Signed manifest
minisign, key 2DC28CC7B26A283B. IndustryCorpus2_DataRater.json. Provenance. Takedown history: none.
Source
huggingface.co/BAAI/IndustryCorpus2_DataRater

Related models

All of alt.freeweights.text →
From: laion <huggingface.co/laion>Subject: a2-rl-e2egit_large-50-32B
From: laion <huggingface.co/laion>Subject: a3-rl-DCAgent_exp_rpt_e2egit-large-15-8B
From: laion <huggingface.co/laion>Subject: a3-rl-DCAgent_exp_rpt_e2egit-v2-10-8B
From: laion <huggingface.co/laion>Subject: a3-rl-DCAgent_exp_rpt_pymethods2test-large-80-8B
From: laion <huggingface.co/laion>Subject: a3-rl-DCAgent_exp_rpt_unitsyn-python-large-10-8B
From: laion <huggingface.co/laion>Subject: a3-rl-DCAgent_mix_h2_language_proportional-65-8B