© 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
    2. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    3. GPT2-Small
    4. Residual Similar CE Local
    5. 6-RES_SCL-AJT
    6. 4838
    Prev
    Next
    INDEX
    Explanations

    government and political positions or titles

    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
    +0.16
    0.6%
    201
    +0.12
    0.5%
    339
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.16
    0.19
    362
    +0.12
    0.11
    604
    +0.12
    0.06
    Negative Logits
    stakes
    -1.01
    hetti
    -0.94
     Gutenberg
    -0.91
    oring
    -0.88
    ĸļ
    -0.87
    ipedia
    -0.87
    oven
    -0.85
    yrs
    -0.85
    obyl
    -0.82
    orem
    -0.82
    POSITIVE LOGITS
     clerks
    1.06
    uty
    1.06
     deputy
    1.05
     assistant
    1.01
     adjunct
    0.95
     sher
    0.94
     sheriff
    0.92
     director
    0.92
     chief
    0.86
     Assistant
    0.86
    Activations Density 6.831%

    No Known Activations