© 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. 3-RES-JB
    11. 17618
    Prev
    Next
    INDEX
    Explanations

    words related to geographical locations, particularly focusing on the country Serbia

    oai_token-act-pair · gpt-3.5-turbo

    mentions of Serbia and related contextual terms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    WARD
    -0.83
    encers
    -0.80
    ITNESS
    -0.78
    enegger
    -0.77
    ership
    -0.77
    ointed
    -0.73
    ORGE
    -0.73
    haw
    -0.72
    aver
    -0.72
    aday
    -0.71
    POSITIVE LOGITS
     Serbia
    1.16
    Ń·
    1.00
     Bosnia
    0.95
     Serbian
    0.93
     Alban
    0.90
     Herz
    0.84
     Ukrain
    0.83
     Slav
    0.80
     Congo
    0.80
     Monteneg
    0.80
    Activations Density 0.014%

    No Known Activations