© 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. 35202
    Prev
    Next
    INDEX
    Explanations

    legal and judicial terms

    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
    +0.19
    0.8%
    631
    +0.17
    0.7%
    481
    +0.14
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.19
    0.28
    452
    +0.17
    0.09
    33
    +0.14
    0.09
    Negative Logits
    ãĤ¦
    -0.79
    ãĥ¡
    -0.76
     weap
    -0.75
     donation
    -0.75
    ontent
    -0.74
    sold
    -0.72
     2030
    -0.72
    ombies
    -0.72
     selves
    -0.71
     broch
    -0.69
    POSITIVE LOGITS
     justices
    1.23
     rulings
    1.12
     upholding
    1.10
     acquitted
    1.10
    Judge
    1.06
     verdict
    1.05
     convict
    1.04
     courtroom
    1.04
     Scalia
    1.04
     Judges
    1.02
    Activations Density 17.315%

    No Known Activations