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. 18067
    Prev
    Next
    INDEX
    Explanations

    names of public figures and politically charged terms

    oai_token-act-pair · gpt-3.5-turbo

    references to specific individuals associated with conspiracy theories and political commentary, particularly Alex Jones and Glenn Beck

    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
     warr
    -0.77
    iour
    -0.74
    ahime
    -0.73
     Thames
    -0.72
     captcha
    -0.70
     concess
    -0.70
     Romeo
    -0.70
     Jagu
    -0.69
    irements
    -0.67
     Vim
    -0.66
    POSITIVE LOGITS
    ervative
    1.17
    ervatives
    1.03
    icist
    0.88
     disinformation
    0.88
    ischer
    0.85
     adherents
    0.82
     propaganda
    0.81
     indoctr
    0.79
     misinformation
    0.79
     hypoc
    0.78
    Activations Density 0.111%

    No Known Activations