© 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. 21034
    Prev
    Next
    INDEX
    Explanations

    adjectives related to the way something is done or the style in which something is performed

    oai_token-act-pair · gpt-3.5-turbo

    terms that describe actions or events in a particular style or way

    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
    addons
    -0.71
    raltar
    -0.67
     oun
    -0.65
    algia
    -0.63
     patriarch
    -0.60
    OTS
    -0.57
    Redd
    -0.57
    reditary
    -0.56
    arus
    -0.56
    umbn
    -0.55
    POSITIVE LOGITS
     reminiscent
    0.84
     resembling
    0.82
     imaginable
    0.78
     (<
    0.76
     thanks
    0.74
     ILCS
    0.73
     depending
    0.72
    .(
    0.71
    ,,,,
    0.71
     throughout
    0.69
    Activations Density 0.255%

    No Known Activations