© 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 Local
    9. 6-RES_SCL-AJT
    11. 35590
    Prev
    Next
    INDEX
    Explanations

    medical terms and technical information related to equipment and procedures, and concern about the lack of oversight in these matters

    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
    +0.25
    1.0%
    735
    +0.14
    0.6%
    195
    +0.13
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    102
    +0.25
    0.29
    55
    +0.14
    0.24
    566
    +0.13
    0.17
    Negative Logits
    ramid
    -1.02
    ongyang
    -0.79
    insula
    -0.74
    imation
    -0.73
    isson
    -0.72
    PsyNetMessage
    -0.72
     Relief
    -0.71
     Tribune
    -0.70
    ãĥ¼ãĥ«
    -0.69
    gee
    -0.69
    POSITIVE LOGITS
     unchanged
    1.35
     unanswered
    1.20
     intact
    1.19
     undecided
    1.08
     undefeated
    1.08
     unaffected
    1.07
     untouched
    1.04
     unsolved
    1.03
     steadfast
    1.01
     afloat
    0.97
    Activations Density 9.218%

    No Known Activations