Skip to content

EthenEthenEthen

LoginPlatformTry Ethen

Open Source Model Profile · barc0

Llama-3.1-ARC-Potpourri-Induction-8B

Llama-3.1-ARC-Potpourri-Induction-8B is an 8.03B-parameter Llama text-generation fine-tune from barc0. Its model card documents induction-dataset training on a Llama-3.1-8B-Instruct base.

Publisher
barc0
Task
text-generation
Model type
llama
License
llama3.1
Library
transformers
Publication status
Accepted · not indexed

Model overview

Llama-3.1-ARC-Potpourri-Induction-8B is published by barc0 as a text-generation fine-tune. The captured configuration identifies LlamaForCausalLM with model type llama, and Safetensors metadata reports 8,030,261,248 parameters. According to the model card, it is a fine-tuned version of Meta-Llama-3.1-8B-Instruct on barc0 induction datasets, with llama3.1 recorded as the license.

Recorded capabilities

Documented Llama-3.1-8B-Instruct base

Hub tags and the model card both identify meta-llama/Llama-3.1-8B-Instruct as the base and fine-tune source.

ARC induction focus

According to the model card, training uses induction datasets with an ARC public-evaluation grid example converted to chat format.

Llama 8.03B configuration

Captured config identifies LlamaForCausalLM and llama, with Safetensors metadata reporting 8,030,261,248 parameters.

Reported loss and hyperparameters

The model card reports evaluation loss of 0.2709 with 2-epoch training, learning rate 1e-05, and Transformers and PyTorch versions.

Use cases in the source record

  • Induction-style grid-reasoning experiments using the documented Llama-3.1 instruct template and ARC example format.
  • Conversational text-generation workflows using the Transformers stack recorded for this Llama model.

Limitations and unknowns

  • No benchmark results beyond the publisher-reported validation loss were extracted.
  • No context-window value, hardware requirement, or current deployment detail was extracted.
  • Provider state is historical snapshot data, not independently refreshed current availability.
  • Base-model, dataset, and training details come from the publisher model card and were not independently verified by Ethen.

Source and provenance

Source: barc0/Llama-3.1-ARC-Potpourri-Induction-8B

Captured: Unknown. Processed: 2026-09-07T19:34:40.815978+00:00.

l3.1-8b-inst-fft-induction-barc-heavy-200k-old-200k-lr1e-5-ep2 This model is a fine-tuned version of meta-llama/Meta-Llama-3.1-8B-Instruct on the barc0/induction_heavy_100k_jsonl, the barc0/induction_heavy_suggestfunction_100k_jsonl, the barc0/induction_100k-gpt4-description-gpt4omini-code_generated_problems_messages_format_0.3 and the barc0/induction_100k_gpt4o-mini_generated_problems_seed100.jsonl_messages_format_0.3 datasets. It achieves the following results on the evaluation set: Loss: 0.2709 Prompt Format We follow Llama-3.1 instruct template. For example, the ARC public evaluation problem 62ab2642 is converted to [{"role": "s…

F001F002F003F004F005F006F007F008F009F010F011F012F013F014F015F016