© 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. 31598
    Prev
    Next
    INDEX
    Explanations

    references to specific names or nouns, possibly related to individuals or entities

    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
    +0.20
    0.9%
    659
    +0.14
    0.6%
    232
    +0.13
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.20
    0.14
    343
    +0.14
    0.08
    232
    +0.13
    0.08
    Negative Logits
    manship
    -0.88
     CPC
    -0.83
    lessness
    -0.80
     ALE
    -0.75
    ãĥĨãĤ£
    -0.72
    IBLE
    -0.72
    flies
    -0.72
     IMAGES
    -0.71
     Cerberus
    -0.70
    20439
    -0.70
    POSITIVE LOGITS
    pper
    1.22
    utsch
    1.17
    ogie
    1.12
    oser
    1.12
    zie
    1.10
    acher
    1.10
    ogle
    1.09
    zan
    1.07
    ogh
    1.06
    isel
    1.05
    Activations Density 6.757%

    No Known Activations