© 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. 14974
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "Savannah"

    oai_token-act-pair · gpt-3.5-turbo

    references to specific names, particularly surnames

    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
    paper
    -0.77
    portation
    -0.76
    ERAL
    -0.68
    pole
    -0.67
    ETH
    -0.66
    lator
    -0.66
     coordination
    -0.64
     fiber
    -0.63
     buckle
    -0.63
     labor
    -0.62
    POSITIVE LOGITS
     Sav
    3.96
    Sav
    2.53
    sav
    1.69
     Savior
    1.52
     sav
    1.39
     Kap
    1.15
     Save
    1.12
     Lov
    1.11
     Salv
    1.05
     Sab
    1.04
    Activations Density 0.021%

    No Known Activations