© 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 E2E Recon
    9. 6-RES_SCEFR-AJT
    11. 328
    Prev
    Next
    INDEX
    Explanations

    technical details and guidelines in a text, potentially related to software development or hardware specifications

    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
    651
    +0.15
    0.5%
    296
    +0.14
    0.5%
    182
    +0.14
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    651
    +0.15
    0.05
    460
    +0.14
    0.05
    300
    +0.14
    0.05
    Negative Logits
    MET
    -0.92
    gee
    -0.85
    english
    -0.85
    MON
    -0.85
    cause
    -0.84
    ctors
    -0.84
    emale
    -0.83
    lde
    -0.82
    tty
    -0.81
    gram
    -0.80
    POSITIVE LOGITS
     onwards
    1.09
    points
    1.08
     onward
    1.03
    abouts
    1.01
    manship
    1.00
    ebin
    1.00
    point
    0.97
    SPONSORED
    0.93
    Ħ¢
    0.92
     iteration
    0.87
    Activations Density 0.575%

    No Known Activations