My pretrained LMs on FineWeb datasets - part of my TensorFlow Model Garden LMs project
Stefan Schweter PRO
stefan-it
AI & ML interests
Flair Library 💕, NER & PoS Tagging, LM Pretraining (mostly encoder-only & encoder-decoder), Historical Language Models, German Language Models, Bavarian NLP 🥨
Recent Activity
liked
a dataset 5 days ago
windprak/steuerllm_instruct_dataset reacted
to
SeaWolf-AI's
post with 🔥 5 days ago
Do Bubbles Form When Tens of Thousands of AIs Simulate Capitalism?
We gave LLMs autonomous trading over 30 real tickers at 100x leverage. All went bankrupt in 30 minutes from hallucination. This spawned FINAL Bench (first metacognition benchmark) and AI NPC Trading Arena — tens of thousands of metacognition-equipped AI agents competing under capitalist rules. Humans can only watch.
Live Demo: https://huggingface.co/spaces/Heartsync/Prompt-Dump
Article: https://huggingface.co/blog/FINAL-Bench/pumpdump
NPCs form a society: 3-tier memory, self-modifying parameters, mutual criticism, strategy propagation, and a virtual SEC enforcing fines every 20 minutes. Every trade passes 4-stage verification including Brave Search fact-check. FINAL Bench confirmed across 9 SOTA models that AI can say "I might be wrong" (MA 0.694) but cannot actually fix errors (ER 0.302).
Six findings: Bubbles form naturally through knowledge transfer and swarm herding. Identical NPCs diverge irreversibly from their first three trades. Metacognition blocks individual hallucination but not collective herding — this is the key finding. Information asymmetry solidifies hierarchy. Fraud and regulation co-evolve. Criticism improves returns.
Individual intelligence does not guarantee collective intelligence.
Dataset & Paper:
https://huggingface.co/datasets/FINAL-Bench/Metacognitive liked
a dataset 5 days ago
castorini/NanoKnow-Fineweb-Edu-Index Organizations
Fined-Tuned NER Models on Historical NER Datasets (HIPE-2022) with Flair and hmTEAMS as backbone LM
Fined-Tuned NER Models on Historical NER Datasets (HIPE-2022) with Flair and hmBERT as backbone LM
-
hmbyt5-preliminary/flair-hipe-2022-ajmc-de
Token Classification • Updated • 2 -
hmbyt5-preliminary/flair-hipe-2022-ajmc-en
Token Classification • Updated -
hmbyt5-preliminary/flair-hipe-2022-ajmc-fr
Token Classification • Updated • 1 -
hmbyt5-preliminary/flair-hipe-2022-newseye-de
Token Classification • Updated • 1
Fined-Tuned NER Models on Historical NER Datasets (HIPE-2022) with Flair and hmBERT Tiny as backbone LM
My pretrained Language Models for Turkish
My fine-tuned NER models for Georgian
-
stefan-it/autotrain-flair-georgian-ner-xlm_r_large-bs4-e10-lr5e-06-1
Token Classification • Updated • 18 -
stefan-it/autotrain-flair-georgian-ner-xlm_r_large-bs4-e10-lr5e-06-2
Token Classification • Updated -
stefan-it/autotrain-flair-georgian-ner-xlm_r_large-bs4-e10-lr5e-06-3
Token Classification • Updated -
stefan-it/autotrain-flair-georgian-ner-xlm_r_large-bs4-e10-lr5e-06-4
Token Classification • Updated • 3
My fine-tuned Flair NER models on CleanCoNLL dataset (with different seeds)
A Collection of Historical Multilingual Language Models
-
dbmdz/bert-base-historic-multilingual-cased
Fill-Mask • 0.1B • Updated • 456 • 8 -
dbmdz/bert-base-historic-multilingual-64k-td-cased
Fill-Mask • 0.1B • Updated • 13 • 2 -
hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax
Updated • 1 -
hmteams/teams-base-historic-multilingual-discriminator
0.1B • Updated • 5
Fined-Tuned NER Models on Historical NER Datasets (HIPE-2022) with Flair and hmBERT as backbone LM
Fine-Tuned Flair Models on German MobIE Dataset using 🤗 AutoTrain SpaceRunner
-
stefan-it/autotrain-flair-mobie-gbert_base-bs16-e10-lr5e-05-2
Token Classification • Updated • 3 -
stefan-it/autotrain-flair-mobie-gbert_base-bs16-e10-lr3e-05-3
Token Classification • Updated • 6 -
stefan-it/autotrain-flair-mobie-gbert_base-bs16-e10-lr5e-05-5
Token Classification • Updated • 2 -
stefan-it/autotrain-flair-mobie-gbert_base-bs16-e10-lr5e-05-3
Token Classification • Updated • 2 • 1
Fined-Tuned NER Models on Historical NER Datasets (HIPE-2022) with Flair and hmBERT 64k as backbone LM
Collection of Microsoft Papers with no code/data release
-
MEGA: Multilingual Evaluation of Generative AI
Paper • 2303.12528 • Published -
MEGAVERSE: Benchmarking Large Language Models Across Languages, Modalities, Models and Tasks
Paper • 2311.07463 • Published • 15 -
Kosmos-2.5: A Multimodal Literate Model
Paper • 2309.11419 • Published • 56 -
A Unified View of Masked Image Modeling
Paper • 2210.10615 • Published
My fine-tuned Flair models on CO-FUN NER Dataset
My trained xLSTM LMs (under development)
My pretrained LMs on FineWeb datasets - part of my TensorFlow Model Garden LMs project
A Collection of Historical Multilingual Language Models
-
dbmdz/bert-base-historic-multilingual-cased
Fill-Mask • 0.1B • Updated • 456 • 8 -
dbmdz/bert-base-historic-multilingual-64k-td-cased
Fill-Mask • 0.1B • Updated • 13 • 2 -
hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax
Updated • 1 -
hmteams/teams-base-historic-multilingual-discriminator
0.1B • Updated • 5
Fined-Tuned NER Models on Historical NER Datasets (HIPE-2022) with Flair and hmTEAMS as backbone LM
Fined-Tuned NER Models on Historical NER Datasets (HIPE-2022) with Flair and hmBERT as backbone LM
Fined-Tuned NER Models on Historical NER Datasets (HIPE-2022) with Flair and hmBERT as backbone LM
-
hmbyt5-preliminary/flair-hipe-2022-ajmc-de
Token Classification • Updated • 2 -
hmbyt5-preliminary/flair-hipe-2022-ajmc-en
Token Classification • Updated -
hmbyt5-preliminary/flair-hipe-2022-ajmc-fr
Token Classification • Updated • 1 -
hmbyt5-preliminary/flair-hipe-2022-newseye-de
Token Classification • Updated • 1
Fine-Tuned Flair Models on German MobIE Dataset using 🤗 AutoTrain SpaceRunner
-
stefan-it/autotrain-flair-mobie-gbert_base-bs16-e10-lr5e-05-2
Token Classification • Updated • 3 -
stefan-it/autotrain-flair-mobie-gbert_base-bs16-e10-lr3e-05-3
Token Classification • Updated • 6 -
stefan-it/autotrain-flair-mobie-gbert_base-bs16-e10-lr5e-05-5
Token Classification • Updated • 2 -
stefan-it/autotrain-flair-mobie-gbert_base-bs16-e10-lr5e-05-3
Token Classification • Updated • 2 • 1
Fined-Tuned NER Models on Historical NER Datasets (HIPE-2022) with Flair and hmBERT Tiny as backbone LM
Fined-Tuned NER Models on Historical NER Datasets (HIPE-2022) with Flair and hmBERT 64k as backbone LM
My pretrained Language Models for Turkish
Collection of Microsoft Papers with no code/data release
-
MEGA: Multilingual Evaluation of Generative AI
Paper • 2303.12528 • Published -
MEGAVERSE: Benchmarking Large Language Models Across Languages, Modalities, Models and Tasks
Paper • 2311.07463 • Published • 15 -
Kosmos-2.5: A Multimodal Literate Model
Paper • 2309.11419 • Published • 56 -
A Unified View of Masked Image Modeling
Paper • 2210.10615 • Published
My fine-tuned NER models for Georgian
-
stefan-it/autotrain-flair-georgian-ner-xlm_r_large-bs4-e10-lr5e-06-1
Token Classification • Updated • 18 -
stefan-it/autotrain-flair-georgian-ner-xlm_r_large-bs4-e10-lr5e-06-2
Token Classification • Updated -
stefan-it/autotrain-flair-georgian-ner-xlm_r_large-bs4-e10-lr5e-06-3
Token Classification • Updated -
stefan-it/autotrain-flair-georgian-ner-xlm_r_large-bs4-e10-lr5e-06-4
Token Classification • Updated • 3
My fine-tuned Flair models on CO-FUN NER Dataset
My fine-tuned Flair NER models on CleanCoNLL dataset (with different seeds)
My trained xLSTM LMs (under development)