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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsBlogSlackPrivacy & 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. 7-RES-JB
    6. 2567
    Prev
    Next
    INDEX
    Explanations

    names of individuals related to controversial or scandalous events

    oai_token-act-pair · gpt-3.5-turbo

    references to accusations and incidents related to sexual misconduct

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 7-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    7
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    OTAL
    -0.81
    arily
    -0.81
    IPS
    -0.77
    istically
    -0.74
    icio
    -0.73
    REE
    -0.70
    AY
    -0.70
    abel
    -0.70
    OTS
    -0.69
    alist
    -0.68
    POSITIVE LOGITS
     allegations
    0.99
     accus
    0.90
     raped
    0.88
     accuser
    0.86
     Weinstein
    0.85
     grop
    0.84
    Sexual
    0.83
     accusations
    0.80
     sexually
    0.78
     rape
    0.77
    Activations Density 0.079%

    No Known Activations