© 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 EvalsExportsAPI 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. 3-RES-JB
    11. 4286
    Prev
    Next
    INDEX
    Explanations

    phrases related to actions or events that draw attention

    oai_token-act-pair · gpt-3.5-turbo

    forms of the word "ballot" and related variations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Stew
    -0.76
    BIL
    -0.68
    roma
    -0.68
    RP
    -0.67
    alph
    -0.66
    existent
    -0.66
    iae
    -0.64
    emetery
    -0.64
    bern
    -0.64
    CV
    -0.64
    POSITIVE LOGITS
     Babe
    0.68
    pedia
    0.64
     innocence
    0.63
     Missile
    0.61
    hatt
    0.61
    dash
    0.61
    hearted
    0.60
     ecstasy
    0.60
     BALL
    0.59
     headlights
    0.59
    Activations Density 0.218%

    No Known Activations