© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 31474
    Prev
    Next
    INDEX
    Explanations

    text related to mathematical or scientific terms and definitions

    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
    87
    +0.36
    3.5%
    266
    +0.28
    2.8%
    447
    +0.14
    1.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    480
    +0.36
    -0.08
    64
    +0.28
    -0.46
    87
    +0.14
    -0.27
    Negative Logits
    20439
    -0.32
     mathemat
    -0.26
    OTA
    -0.26
    ãĥīãĥ©ãĤ´ãĥ³
    -0.25
    Ble
    -0.25
    doi
    -0.25
    CLASSIFIED
    -0.24
    Haw
    -0.23
    ©¶æ¥µ
    -0.23
     Sebastian
    -0.23
    POSITIVE LOGITS
    roma
    0.30
    ulent
    0.29
    chuk
    0.27
    senal
    0.27
    foundland
    0.27
    ommel
    0.26
    ipedia
    0.26
    uterte
    0.26
    rosis
    0.26
    uncle
    0.26
    Activations Density 38.576%

    No Known Activations