>, <|input_end|>). - **Output:** Produces a comma-separated list of tags in the correct block, often in alphabetical order. - **Intended Use:** Tag suggestion/completion for Danbooru-style image datasets or tools. - **Limitations:** Does not support natural language input or flexible prompt formats; only works with the specified tag structure. - **License:** Apache-2.0 **Example Prompt:** ``` <|bos|>rating:sfw, rating:generaloriginal<|long|>1girl<|input_end|> ``` **Example Output:** ``` rating:sfw, rating:general, 1girl, ahoge, braid, closed eyes, ... (and more tags) ``` **Best For:** Developers building Danbooru taggers, dataset annotators, or similar tools needing automated, structured tag generation.3e:T44c,Certainly! Here’s a summary of the model **kanishka/smolm-autoreg-bpe-counterfactual-babylm-indef_articles_with_pl_nouns-removal-3e-3**: - **Type:** Causal Language Model (for text generation) - **Architecture:** OPT (Open Pretrained Transformer) for Causal Language Modeling - **Parameters:** ~98 million (97,838,592) - **Precision:** 32-bit - **Tokenizer:** Byte-Pair Encoding (BPE) with special tokens for BOS, EOS, PAD, and UNK - **Hardware Requirements:** 1 GB RAM, 1 GB disk space, suitable for micro instances (e.g., g4dn.xlarge) - **Training Objective:** The model is fine-tuned to remove indefinite articles ("a", "an") before plural nouns, as indicated by the name. This likely means it was trained on a counterfactual version of the BabyLM dataset where such articles were removed. - **Intended Use:** Text generation, especially for tasks or research involving grammaticality, article usage, or counterfactual language modeling. **Note:** There are currently no further fine-tuned variants of this model available. If you need more technical details or usage examples, let me know!3f:T607,Certainly! Here’s a summary of the model iproskurina/opt-6.7b-GPTQ-4bit-g128: **Model Name:** iproskurina/opt-6.7b-GPTQ-4bit-g128 **Base Mod