© 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. 4-RES-JB
    11. 6968
    Prev
    Next
    INDEX
    Explanations

    phrases related to unexpected events or actions

    oai_token-act-pair · gpt-3.5-turbo

    phrases that describe the concept of being without warning or permission

    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
    soDeliveryDate
    -0.81
    DragonMagazine
    -0.78
    nai
    -0.76
    姫
    -0.73
    ashtra
    -0.73
    estate
    -0.72
     Christy
    -0.70
    odynam
    -0.66
    geist
    -0.63
    uesday
    -0.63
    POSITIVE LOGITS
     whatsoever
    0.91
     interruption
    0.83
     slightest
    0.78
     knowledge
    0.75
     illumination
    0.74
     nor
    0.74
     touching
    0.70
     mention
    0.70
     fuss
    0.68
     distractions
    0.68
    Activations Density 0.176%

    No Known Activations