© 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. 2-RES-JB
    11. 21795
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "ape" within the text, focusing on instances where the word "ape" is part of another word

    oai_token-act-pair · gpt-3.5-turbo

    instances of the substring "ape" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    quickShipAvailable
    -0.75
    sch
    -0.75
    naire
    -0.73
    BIL
    -0.69
    SOURCE
    -0.68
    ship
    -0.65
    Mac
    -0.64
    naires
    -0.63
    sf
    -0.62
    angler
    -0.61
    POSITIVE LOGITS
    utic
    0.94
    etus
    0.94
    ape
    0.92
    utics
    0.86
    yre
    0.83
    oton
    0.76
    uracy
    0.76
    aukee
    0.76
    lla
    0.75
    apers
    0.74
    Activations Density 0.009%

    No Known Activations