© 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. 10991
    Prev
    Next
    INDEX
    Explanations

    phrases related to standing by or supporting someone or something

    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
    •Layer 6 UMAP region: Mostly-local cluster on left - e2e-recon

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    78
    +0.15
    0.5%
    57
    +0.15
    0.4%
    393
    +0.15
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    237
    +0.15
    0.05
    238
    +0.15
    0.05
    29
    +0.15
    0.06
    Negative Logits
    ibur
    -1.18
    ãĥ´ãĤ¡
    -1.16
    ayne
    -1.12
    alde
    -1.10
    mble
    -1.06
     comprom
    -1.06
    llor
    -1.04
    inary
    -1.04
    hement
    -1.02
    reement
    -1.01
    POSITIVE LOGITS
    products
    1.61
    laws
    1.40
    product
    1.40
     virtue
    1.21
     extension
    1.16
     leaps
    1.16
     omission
    1.12
     unnoticed
    1.10
     proxy
    1.08
     nature
    1.03
    Activations Density 0.970%

    No Known Activations