© 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
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 10-RES-JB
    11. 3099
    Prev
    Next
    INDEX
    Explanations

    the word "Cap" with varying levels of importance or relevance

    oai_token-act-pair · gpt-3.5-turbo

    capitalized words or terms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    hower
    -0.91
    ĪĴ
    -0.77
    IGHTS
    -0.74
     silence
    -0.71
    ¿½
    -0.71
     anten
    -0.65
    gaard
    -0.64
     Nare
    -0.63
     darts
    -0.63
     bye
    -0.61
    POSITIVE LOGITS
    itol
    1.42
    itals
    1.27
    acity
    1.27
    rice
    1.27
    abilities
    1.26
    rices
    1.25
    illary
    1.21
    uchin
    1.16
    rylic
    1.11
    itated
    1.09
    Activations Density 0.015%

    No Known Activations