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. 21149
    Prev
    Next
    INDEX
    Explanations

    mentions of a specific news outlet

    oai_token-act-pair · gpt-3.5-turbo

    references to the news outlet Politico

    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
    alogue
    -0.73
    unts
    -0.72
    venge
    -0.71
    unal
    -0.69
     captcha
    -0.69
     restitution
    -0.68
    ĸļ
    -0.67
    araoh
    -0.67
     awoken
    -0.66
    thro
    -0.65
    POSITIVE LOGITS
     Magazine
    1.38
     magazine
    1.10
    Magazine
    1.08
     Editors
    0.95
     Weekly
    0.93
     Digest
    0.92
     editors
    0.92
     News
    0.91
     Coverage
    0.88
     Insider
    0.87
    Activations Density 0.060%

    No Known Activations