© 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. 13060
    Prev
    Next
    INDEX
    Explanations

    words and suffixes related to abstract qualities or states

    oai_token-act-pair · gpt-3.5-turbo

    words and concepts related to different forms of "ness" or states of being

    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
    bsite
    -0.71
    inav
    -0.68
    vice
    -0.68
    umm
    -0.66
    herty
    -0.66
    nesium
    -0.65
    opa
    -0.63
    OY
    -0.62
    jam
    -0.62
    bold
    -0.62
    POSITIVE LOGITS
     quot
    0.94
    iness
    0.94
     bordering
    0.87
    lessness
    0.81
     perv
    0.80
     incarn
    0.80
    ously
    0.80
     Principle
    0.79
    phal
    0.76
    uality
    0.76
    Activations Density 0.172%

    No Known Activations