© 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. 3885
    Prev
    Next
    INDEX
    Explanations

    specific values or statistics mentioned in 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
    +0.46
    39.2%
    138
    +0.13
    10.8%
    64
    +0.00
    0.1%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    138
    +0.46
    1.00
    447
    +0.13
    0.99
    64
    +0.00
    -0.01
    Negative Logits
     Azerb
    -0.02
    ij士
    -0.02
    oÄŁ
    -0.02
    elsius
    -0.02
    é¾
    -0.02
    Þ
    -0.02
     guiActiveUn
    -0.02
    ļéĨĴ
    -0.02
    ñ
    -0.02
    ÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤ
    -0.02
    POSITIVE LOGITS
     the
    0.03
    ,
    0.02
     and
    0.02
     in
    0.02
    ↵
    0.02
    .
    0.02
     a
    0.02
     to
    0.02
     of
    0.02
     is
    0.02
    Activations Density 18.012%

    No Known Activations