© 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. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    5. GPT2-Small
    7. Residual Similar CE Local
    9. 6-RES_SCL-AJT
    11. 4981
    Prev
    Next
    INDEX
    Explanations

    descriptions of political events and processes, particularly related to election primaries

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scl-ajt/6-res_scl-ajt
    Prompts (Dashboard)
    12,288 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    46,080
    Data Type
    torch.float32
    Hook Point
    blocks.6.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    apollo-research/Skylion007-openwebtext-tokenizer-gpt2
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    447
    +0.19
    0.9%
    288
    +0.12
    0.6%
    492
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    257
    +0.19
    0.06
    310
    +0.12
    0.07
    452
    +0.11
    0.05
    Negative Logits
    BG
    -0.70
    dfx
    -0.67
    isine
    -0.66
     Orbital
    -0.61
    uge
    -0.61
    depend
    -0.60
    ug
    -0.60
    relative
    -0.60
    ffect
    -0.59
    usal
    -0.59
    POSITIVE LOGITS
     season
    0.94
     challenger
    0.79
     electorate
    0.79
     challengers
    0.77
     elections
    0.76
     runoff
    0.76
     voters
    0.75
     election
    0.73
     opponents
    0.72
    halla
    0.71
    Activations Density 5.043%

    No Known Activations