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. 8-RES-JB
    6. 6786
    Prev
    Next
    INDEX
    Explanations

    financial and criminal activities, such as payments, bribes, and conspiracies

    oai_token-act-pair · gpt-3.5-turbo

    instances of financial misconduct or illegal activities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ï¸ı
    -0.75
    ĪĴ
    -0.72
    iets
    -0.71
     :(
    -0.69
    annel
    -0.68
     ðŁij
    -0.67
    urally
    -0.67
    eatures
    -0.66
     ðŁĺ
    -0.66
     âľ
    -0.66
    POSITIVE LOGITS
     prostitutes
    1.11
     blackmail
    1.09
     pornographic
    1.08
     extortion
    1.04
     cocaine
    1.02
     bribes
    1.02
     unsuspecting
    0.99
     raping
    0.98
     ransom
    0.98
     prostitute
    0.97
    Activations Density 1.212%

    No Known Activations