Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 45516
    Prev
    Next
    INDEX
    Explanations

    mentions of Congress and legislative actions

    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
    +0.28
    1.4%
    385
    +0.09
    0.4%
    488
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    762
    +0.28
    0.20
    718
    +0.09
    0.16
    566
    +0.08
    0.12
    Negative Logits
     Britann
    -0.59
    jiang
    -0.59
     Vertical
    -0.58
     Fatal
    -0.56
     MM
    -0.56
     9000
    -0.54
     Mighty
    -0.54
     True
    -0.54
    omorph
    -0.53
     Oprah
    -0.53
    POSITIVE LOGITS
    woman
    0.96
    ional
    0.83
     adjourn
    0.78
    member
    0.78
     appropriated
    0.75
     appropri
    0.71
    arians
    0.70
     enacted
    0.70
    members
    0.69
     appropriations
    0.69
    Activations Density 11.588%

    No Known Activations