© 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 EvalsExportsinterp-engineNEWAPI 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. 8078
    Prev
    Next
    INDEX
    Explanations

    words related to countries and territories

    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
    609
    +0.07
    0.5%
    571
    +0.07
    0.5%
    17
    +0.07
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    57
    +0.07
    0.02
    727
    +0.07
    0.02
    534
    +0.07
    0.01
    Negative Logits
    anski
    -0.56
    anian
    -0.52
     Croat
    -0.50
    yip
    -0.49
    icz
    -0.49
    Downloadha
    -0.48
    ruary
    -0.47
    imbabwe
    -0.47
    anni
    -0.47
    ividual
    -0.47
    POSITIVE LOGITS
    ¢
    0.50
    cedes
    0.49
    Mist
    0.48
    Fac
    0.48
    ãģĦ
    0.47
    ãĥ¤
    0.45
    STAR
    0.44
    direction
    0.43
     sole
    0.43
    Cur
    0.43
    Activations Density 0.072%

    No Known Activations