© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 4-RES-JB
    6. 20486
    Prev
    Next
    INDEX
    Explanations

    statements about someone being declared dead

    oai_token-act-pair · gpt-3.5-turbo

    instances of the term "pronounced" often in medical or formal contexts related to death or status updates

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

    No Comments

    Negative Logits
    ionage
    -0.82
     Camer
    -0.76
    bies
    -0.68
    chedel
    -0.67
    allow
    -0.67
    hatt
    -0.67
    committee
    -0.66
    OAD
    -0.66
    aldo
    -0.66
     Clubs
    -0.65
    POSITIVE LOGITS
     pronounced
    1.02
     pronounce
    0.92
     pronunciation
    0.88
    unci
    0.86
     audible
    0.78
    entimes
    0.77
    aneously
    0.77
     pron
    0.75
     conson
    0.74
     deaf
    0.73
    Activations Density 0.011%

    No Known Activations