© 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. 9-RES-JB
    11. 4128
    Prev
    Next
    INDEX
    Explanations

    phrases containing the word "Sel" which could potentially relate to "Diesel," "Joinsel," or "Kuurne-Brussel-Kuurne."

    oai_token-act-pair · gpt-3.5-turbo

    proper nouns and names associated with specific people or entities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    9
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    hered
    -0.66
    ģ
    -0.63
    elf
    -0.60
     cess
    -0.60
     millenn
    -0.59
     nomine
    -0.59
     resilient
    -0.58
     Tune
    -0.58
    ItemTracker
    -0.58
    ULTS
    -0.58
    POSITIVE LOGITS
    anguage
    1.30
    ounge
    1.18
    ibrary
    1.15
    ength
    1.15
    anguages
    1.10
    mans
    1.06
    le
    1.05
    mann
    1.05
    bach
    1.02
    abor
    1.01
    Activations Density 0.088%

    No Known Activations