© 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. 3-RES-JB
    6. 15597
    Prev
    Next
    INDEX
    Explanations

    names or references to individuals named "Grady" with varying levels of intensity of activation

    oai_token-act-pair · gpt-3.5-turbo

    references to individuals named "Grady" and variations of the name within various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    uckland
    -0.72
    UME
    -0.72
    olated
    -0.70
    falls
    -0.67
    bidden
    -0.65
    osion
    -0.65
    ingen
    -0.64
    anamo
    -0.64
    semb
    -0.63
    ModLoader
    -0.63
    POSITIVE LOGITS
    rov
    0.77
    jet
    0.75
    mbol
    0.74
    tsky
    0.68
    ady
    0.68
    hots
    0.68
    illac
    0.68
    ng
    0.67
    stre
    0.66
     GOODMAN
    0.66
    Activations Density 0.010%

    No Known Activations