© 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. 11653
    Prev
    Next
    INDEX
    Explanations

    positive descriptions of material quality and visual aspects like design and art in various contexts, such as in products, games, and covers

    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
    393
    +0.28
    1.0%
    566
    +0.14
    0.5%
    215
    +0.14
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    55
    +0.28
    0.18
    566
    +0.14
    0.16
    102
    +0.14
    0.16
    Negative Logits
    ourt
    -0.92
    inav
    -0.91
    edIn
    -0.90
    commit
    -0.90
    ocobo
    -0.89
    yout
    -0.88
    isco
    -0.88
    uld
    -0.88
    ervatives
    -0.87
    STAR
    -0.86
    POSITIVE LOGITS
     not
    1.15
     supposed
    1.14
    nt
    1.12
     also
    1.12
     hereby
    1.12
     definitely
    1.11
     NOT
    1.06
     interchangeable
    1.05
     unchanged
    1.05
     always
    1.03
    Activations Density 2.805%

    No Known Activations