© 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. 3858
    Prev
    Next
    INDEX
    Explanations

    published content or materials

    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
    +0.26
    1.2%
    288
    +0.13
    0.6%
    183
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.26
    0.27
    385
    +0.13
    0.23
    608
    +0.11
    0.26
    Negative Logits
    llan
    -0.90
    uana
    -0.74
     awakened
    -0.72
    xa
    -0.72
    Ĭ±
    -0.70
    uppet
    -0.70
    pell
    -0.69
    eric
    -0.69
    vette
    -0.68
    aturation
    -0.68
    POSITIVE LOGITS
    lishing
    1.14
    lisher
    1.11
     excerpts
    0.93
    lishes
    0.89
     showc
    0.73
    INC
    0.73
     articles
    0.71
     newsp
    0.71
     Seym
    0.70
     publication
    0.70
    Activations Density 11.419%

    No Known Activations