© 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. 33287
    Prev
    Next
    INDEX
    Explanations

    phrases related to conflicts and military 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
    +0.18
    0.7%
    97
    +0.11
    0.4%
    526
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.18
    0.38
    362
    +0.11
    0.18
    393
    +0.10
    0.34
    Negative Logits
    ç·
    -0.94
    olor
    -0.79
    iterranean
    -0.74
    awan
    -0.73
    loe
    -0.73
    REDACTED
    -0.73
    >[
    -0.72
    women
    -0.71
    æĸ¹
    -0.71
    çĶ°
    -0.70
    POSITIVE LOGITS
     favour
    0.88
     afar
    0.82
     sight
    0.81
     favor
    0.80
     vitro
    0.79
    clusively
    0.73
     spite
    0.73
     senseless
    0.72
     sheer
    0.72
    wards
    0.72
    Activations Density 16.714%

    No Known Activations