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. 1-RES-JB
    6. 1272
    Prev
    Next
    INDEX
    Explanations

    mentions of the term "paed" (related to pedophilia/pedophiles)

    oai_token-act-pair · gpt-3.5-turbo

    terms related to pedophilia and legal implications surrounding it

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    place
    -0.72
    ball
    -0.66
    WM
    -0.66
    bow
    -0.66
    è¦ļéĨĴ
    -0.65
    IPS
    -0.65
    balls
    -0.64
    Beh
    -0.64
    VP
    -0.64
     Irving
    -0.62
    POSITIVE LOGITS
    aed
    1.31
    ternity
    1.04
    ophile
    1.04
    iatrics
    0.93
    amily
    0.93
    aily
    0.91
    iciary
    0.89
    ersen
    0.89
    emy
    0.88
    usa
    0.87
    Activations Density 0.007%

    No Known Activations