© 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. 8-RES-JB
    11. 12303
    Prev
    Next
    INDEX
    Explanations

    phrases related to visual descriptions, specifically focusing on the eyes

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to visual appeal and catchiness

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    illet
    -0.67
    Together
    -0.67
    urdue
    -0.64
    DN
    -0.64
    hement
    -0.61
    eros
    -0.60
    Anth
    -0.59
    minus
    -0.59
     upkeep
    -0.59
     CK
    -0.58
    POSITIVE LOGITS
     beh
    1.04
     ears
    0.88
    ĸļ
    0.82
    itness
    0.81
    catching
    0.76
    SEE
    0.76
    owler
    0.71
     dart
    0.71
    oscope
    0.70
    eyed
    0.70
    Activations Density 0.186%

    No Known Activations