© 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
    2. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    3. GPT2-Small
    4. Residual Similar CE Local
    5. 6-RES_SCL-AJT
    6. 39714
    Prev
    Next
    INDEX
    Explanations

    pairs of words that are somehow connected in a specific context

    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
    +0.23
    1.2%
    316
    +0.09
    0.5%
    418
    +0.09
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.23
    0.19
    385
    +0.09
    0.20
    608
    +0.09
    0.18
    Negative Logits
    uve
    -0.72
    lam
    -0.72
    arer
    -0.66
    bay
    -0.66
    avis
    -0.65
    ker
    -0.61
    cher
    -0.61
    van
    -0.61
    glers
    -0.60
    fitting
    -0.60
    POSITIVE LOGITS
     respective
    0.90
     strengths
    0.77
     halves
    0.72
     sides
    0.71
     merits
    0.70
     branches
    0.69
     genders
    0.68
     administrations
    0.68
     colours
    0.67
     capacities
    0.66
    Activations Density 10.684%

    No Known Activations