About 8.03B parameters
Safetensors metadata reports 8,030,261,248 parameters, or about 8.03B.
Open Source Model Profile · nbeerbower
llama-3-spicy-abliterated-stella-8B
llama-3-spicy-abliterated-stella-8B is an 8.03B-parameter Llama text-generation model from nbeerbower. The model card describes it as a mergekit merge using the Model Stock method.
- Publisher
- nbeerbower
- Task
- text-generation
- Model type
- llama
- License
- llama3
- Library
- transformers
- Publication status
- Approved for indexing
Model overview
nbeerbower publishes llama-3-spicy-abliterated-stella-8B as a text-generation model. Captured config identifies LlamaForCausalLM with a llama model type, and Safetensors metadata reports 8,030,261,248 parameters. Card data records a llama3 license. Hub tags include mergekit and merge. According to the model card, the merge used Model Stock with nbeerbower/llama-3-spicy-8B as a base.
Recorded capabilities
llama3 licensing
Captured metadata records a llama3 license.
Model Stock merge
The model card says this is a mergekit merge using Model Stock, with nbeerbower/llama-3-spicy-8B as the base, and lists nbeerbower/llama-3-stella-8B, saishf/Aura-Uncensored-OAS-8B-L3, and cognitivecomputations/Llama-3-8B-Instruct-abliterated-v2 among merged models.
Card-reported leaderboard figures
The model card lists Open LLM Leaderboard figures including Avg. 69.27, ARC 65.19, HellaSwag 81.98, MMLU 67.88, TruthfulQA 53.65, Winogrande 76.48, and GSM8k 70.43. Those figures are publisher/card claims, not independently verified Ethen results.
Use cases in the source record
- Transformers text-generation and conversational workflows on the tagged Llama causal-LM stack.
Limitations and unknowns
- No Ethen-verified evaluation results were extracted. Leaderboard numbers on the card remain publisher claims.
- Provider state is historical snapshot data, not independently refreshed current availability.
- Merge ingredients, Model Stock method, and filename tokens such as spicy or abliterated are not treated as independently verified capability claims.
Source and provenance
Source: nbeerbower/llama-3-spicy-abliterated-stella-8B
Captured: Unknown. Processed: 2026-09-07T19:34:53.119935+00:00.
llama-3-spicy-abliterated-stella-8B This is a merge of pre-trained language models created using mergekit . Merge Details Merge Method This model was merged using the Model Stock merge method using nbeerbower/llama-3-spicy-8B as a base. Models Merged The following models were included in the merge: nbeerbower/llama-3-stella-8B saishf/Aura-Uncensored-OAS-8B-L3 cognitivecomputations/Llama-3-8B-Instruct-abliterated-v2 Configuration The following YAML configuration was used to produce this model: models: - model: nbeerbower/llama-3-stella-8B - model: saishf/Aura-Uncensored-OAS-8B-L3 - model: cognitivecomputations/Llama-3-8B-Instruct-abl…
F001F002F003F004F005F006F007F008F009F010F011F012F013