© 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 E2E Recon
    9. 6-RES_SCEFR-AJT
    11. 14211
    Prev
    Next
    INDEX
    Explanations

    terms related to vulnerabilities, especially in critical infrastructure and sensitive systems

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scefr-ajt/6-res_scefr-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
    288
    +0.24
    1.0%
    99
    +0.17
    0.7%
    161
    +0.13
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.24
    0.13
    151
    +0.17
    0.10
    417
    +0.13
    0.06
    Negative Logits
    zos
    -0.98
    tein
    -0.86
    icle
    -0.83
     Collider
    -0.82
    aster
    -0.81
    oret
    -0.80
    ove
    -0.80
    inth
    -0.78
    Ħ¢
    -0.77
    rams
    -0.76
    POSITIVE LOGITS
     vulnerable
    1.10
     vulner
    1.05
    ulnerable
    0.99
     vulnerability
    0.95
     vulnerabilities
    0.95
     weaknesses
    0.89
     adolesc
    0.88
    ulnerability
    0.87
    ity
    0.84
     defences
    0.82
    Activations Density 1.929%

    No Known Activations