© 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. 33292
    Prev
    Next
    INDEX
    Explanations

    information leading up to an event or situation

    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
    +0.22
    1.1%
    218
    +0.10
    0.5%
    622
    +0.09
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    240
    +0.22
    0.10
    608
    +0.10
    0.18
    306
    +0.09
    0.12
    Negative Logits
    abase
    -0.64
    uffy
    -0.62
    wagen
    -0.58
     guiIcon
    -0.56
    ifty
    -0.55
    pload
    -0.55
    yo
    -0.55
     ILCS
    -0.54
     beware
    -0.54
    uni
    -0.54
    POSITIVE LOGITS
     Ascension
    0.67
    imester
    0.65
     Election
    0.63
    Sept
    0.63
     dusk
    0.63
     Ramadan
    0.63
     adulthood
    0.62
    stage
    0.61
    mortem
    0.60
     Madness
    0.60
    Activations Density 11.626%

    No Known Activations