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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 32507
    Prev
    Next
    INDEX
    Explanations

    Sentence structures related to laws, regulations, and legal document descriptions

    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
    +0.34
    1.8%
    75
    +0.10
    0.5%
    74
    +0.09
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    608
    +0.34
    0.19
    679
    +0.10
    0.21
    624
    +0.09
    0.07
    Negative Logits
    Mom
    -0.58
    stack
    -0.56
    Sit
    -0.54
     Examiner
    -0.53
    pedia
    -0.53
     detector
    -0.53
     syndrome
    -0.52
    Bed
    -0.52
    Sensor
    -0.50
    score
    -0.50
    POSITIVE LOGITS
    yrinth
    0.62
    actly
    0.60
    hammad
    0.58
     formatting
    0.57
    irtual
    0.54
    itled
    0.53
     deduct
    0.53
     unmarried
    0.53
     foreigners
    0.52
    orpor
    0.52
    Activations Density 14.317%

    No Known Activations