© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 39851
    Prev
    Next
    INDEX
    Explanations

    references to numbers within text

    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
    +0.31
    1.8%
    393
    +0.20
    1.1%
    71
    +0.16
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    393
    +0.31
    0.28
    375
    +0.20
    0.08
    71
    +0.16
    0.04
    Negative Logits
    hesis
    -0.56
    ared
    -0.56
    asion
    -0.55
    ĸļ
    -0.54
    ļéĨĴ
    -0.54
    acus
    -0.54
    acle
    -0.54
    otto
    -0.53
    usp
    -0.53
    =]
    -0.53
    POSITIVE LOGITS
     unanswered
    0.74
     occasions
    0.71
     factors
    0.70
     different
    0.67
     times
    0.66
     reasons
    0.66
     iterations
    0.66
     ways
    0.65
     instances
    0.63
     ailments
    0.63
    Activations Density 10.635%

    No Known Activations