© 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. 3-RES-JB
    11. 20579
    Prev
    Next
    INDEX
    Explanations

    phrases related to cutting, such as "cut," and "cutting edge."

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "cut" and its variations in different 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
    AMP
    -0.76
    KY
    -0.71
    Registered
    -0.70
    agher
    -0.69
    ña
    -0.67
    apo
    -0.67
    BUS
    -0.66
    ESSION
    -0.65
    heid
    -0.64
    igmat
    -0.63
    POSITIVE LOGITS
    scene
    1.18
    scenes
    1.15
    lasses
    1.13
    thro
    1.07
    aneous
    0.98
    lass
    0.97
    backs
    0.94
    cuts
    0.92
    cutting
    0.89
    cut
    0.88
    Activations Density 0.026%

    No Known Activations