© 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. 3110
    Prev
    Next
    INDEX
    Explanations

    the word "Hur" followed by various word endings

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "Hur," likely indicating a focus on various contexts or mentions of "Hurricane."

    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
     Skydragon
    -0.76
    nesota
    -0.72
    é¾įå¥ij士
    -0.66
    OTE
    -0.65
     ANGEL
    -0.64
     Stronghold
    -0.64
     Babel
    -0.62
     Metatron
    -0.60
    volent
    -0.59
    ographically
    -0.59
    POSITIVE LOGITS
    ricane
    1.44
    ricanes
    1.26
    witz
    1.14
    acan
    1.12
    rah
    1.09
    rican
    1.06
    ley
    1.05
    rying
    1.02
    ting
    1.00
    ts
    0.99
    Activations Density 0.032%

    No Known Activations