© 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. 0-RES-JB
    11. 14990
    Prev
    Next
    INDEX
    Explanations

    The neuron is looking for references to the word "sing"

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "Sing" and related phrases or names

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Front
    -0.80
     Leaf
    -0.72
     Slate
    -0.70
     Crest
    -0.69
     temperament
    -0.68
     Café
    -0.67
     Frost
    -0.67
     Division
    -0.67
     Bills
    -0.66
     ensemble
    -0.66
    POSITIVE LOGITS
    Sing
    3.31
    Mal
    1.92
    Cause
    1.72
    cause
    1.51
    Bah
    1.34
    Ring
    1.33
    Sil
    1.26
    Haw
    1.24
    Ali
    1.23
    Que
    1.21
    Activations Density 0.037%

    No Known Activations