© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 11-RES-JB
    11. 12390
    Prev
    Next
    INDEX
    Explanations

    phrases prompting action or involvement

    oai_token-act-pair · gpt-3.5-turbo

    calls to action or prompts encouraging engagement or participation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ¥µ
    -0.67
    vale
    -0.67
    onement
    -0.65
    essee
    -0.64
     Defeat
    -0.60
    odore
    -0.60
    velt
    -0.60
    deen
    -0.60
    ourgeois
    -0.60
    uably
    -0.59
    POSITIVE LOGITS
     acquainted
    1.40
     started
    1.30
    started
    1.21
     involved
    1.12
     Started
    1.11
     notified
    1.04
     rid
    1.00
     ready
    0.95
     immersed
    0.92
     underway
    0.92
    Activations Density 0.099%

    No Known Activations