© 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. 3915
    Prev
    Next
    INDEX
    Explanations

    mentions of specific people or entities in a context related to organized events or matches

    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
    +0.47
    2.5%
    640
    +0.34
    1.8%
    160
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.47
    0.44
    160
    +0.34
    0.29
    745
    +0.12
    -0.12
    Negative Logits
     union
    -0.69
     deflation
    -0.67
     billing
    -0.64
     spir
    -0.64
     covenant
    -0.63
     tackling
    -0.62
    uca
    -0.61
     ticket
    -0.61
     negotiating
    -0.60
     scaling
    -0.60
    POSITIVE LOGITS
    Conclusion
    1.30
    Advertisements
    1.19
    Trivia
    1.18
    âĵĺ
    1.10
    Lastly
    1.07
    Quotes
    1.07
    --------------------------------------------------------
    1.04
    CONCLUS
    1.03
    References
    1.01
    Recommended
    1.00
    Activations Density 14.767%

    No Known Activations