© 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. 1-RES-JB
    11. 13796
    Prev
    Next
    INDEX
    Explanations

    mentions of physical dirt or a negative connotation related to wrongdoing

    oai_token-act-pair · gpt-3.5-turbo

    mentions of "dirt" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    itals
    -0.79
    eer
    -0.74
    alone
    -0.71
    ommod
    -0.68
    inel
    -0.65
    inion
    -0.65
    pha
    -0.65
    emic
    -0.64
    ONT
    -0.64
    ofi
    -0.64
    POSITIVE LOGITS
     dirt
    1.32
     Dirt
    1.14
    bags
    1.08
    bag
    1.00
    bike
    0.88
    iest
    0.79
    gew
    0.78
    iless
    0.78
     mole
    0.77
    sie
    0.75
    Activations Density 0.005%

    No Known Activations