© 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. 9-RES-JB
    11. 19448
    Prev
    Next
    INDEX
    Explanations

    the word "call" followed by a noun or adjective to describe something or give it a label

    oai_token-act-pair · gpt-3.5-turbo

    phrases that involve calling something by a specific name or descriptor

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    9
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ashington
    -0.73
    hung
    -0.73
    hement
    -0.69
    idth
    -0.69
    iston
    -0.65
    icka
    -0.64
    chief
    -0.61
    itely
    -0.61
    Items
    -0.60
    umber
    -0.60
    POSITIVE LOGITS
     qu
    0.77
     heresy
    0.73
     bluff
    0.72
     "#
    0.72
     treason
    0.71
    selves
    0.68
     hypocrisy
    0.67
     parity
    0.66
    self
    0.66
     arrogance
    0.66
    Activations Density 0.044%

    No Known Activations