© 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. 0-RES-JB
    11. 15270
    Prev
    Next
    INDEX
    Explanations

    references to invisibility or things that are hidden from sight

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "invisible" or its variations, indicating a focus on themes of invisibility

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    jong
    -0.84
    ï¸ı
    -0.74
    brook
    -0.67
     Cour
    -0.67
     Daly
    -0.65
     Catalonia
    -0.64
    gow
    -0.64
     Cox
    -0.64
    swick
    -0.64
    utions
    -0.63
    POSITIVE LOGITS
     invisible
    3.67
     Invisible
    2.48
     unseen
    2.36
     invis
    2.19
    visible
    1.78
     imaginary
    1.52
     intangible
    1.49
     undet
    1.48
     invincible
    1.48
    hidden
    1.38
    Activations Density 0.019%

    No Known Activations