© 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 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. 33418
    Prev
    Next
    INDEX
    Explanations

    phrases related to terrorists attacks, shootings, and extremist activities

    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
    346
    +0.11
    0.5%
    121
    +0.11
    0.5%
    47
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.11
    0.12
    288
    +0.11
    0.20
    343
    +0.11
    0.10
    Negative Logits
    dit
    -0.88
     Habit
    -0.77
     Protoss
    -0.74
     Dhabi
    -0.73
     Scotch
    -0.73
    utical
    -0.72
    Champ
    -0.70
     Hawai
    -0.70
     nep
    -0.68
    Ancient
    -0.68
    POSITIVE LOGITS
     victims
    1.15
     shootings
    1.11
     shooter
    1.11
     bombing
    1.10
     gunman
    1.07
     rampage
    1.05
     massacre
    1.04
     gunned
    1.03
    arnaev
    1.02
     bombings
    1.02
    Activations Density 5.875%

    No Known Activations