Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 4-RES-JB
    6. 15074
    Prev
    Next
    INDEX
    Explanations

    explicit sexual content and related actions

    oai_token-act-pair · gpt-3.5-turbo

    sexual content and themes, particularly those involving explicit scenes and behaviors

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 4-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.4.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    4
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    é¾
    -0.82
    ETF
    -0.81
     Kessler
    -0.75
     Canaver
    -0.75
    IELD
    -0.74
    £ı
    -0.73
    REE
    -0.73
     condem
    -0.72
    Weather
    -0.71
    é¾įå¥ij士
    -0.71
    POSITIVE LOGITS
     erotic
    1.18
     slut
    1.16
     nude
    1.13
     sex
    1.12
     sexy
    1.08
     nipples
    1.07
     boobs
    1.06
     panties
    1.06
     kissing
    1.05
     pussy
    1.05
    Activations Density 0.409%

    No Known Activations