© 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. 32392
    Prev
    Next
    INDEX
    Explanations

    phrases related to education and academic critiques

    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
    +0.29
    1.8%
    7
    +0.08
    0.5%
    714
    +0.07
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    393
    +0.29
    0.35
    102
    +0.08
    0.21
    566
    +0.07
    0.14
    Negative Logits
    luaj
    -0.50
     sidx
    -0.47
    selage
    -0.47
    çīĪ
    -0.47
     Lauder
    -0.46
    album
    -0.46
    onnaissance
    -0.45
    Ur
    -0.45
    OOD
    -0.44
    ysis
    -0.44
    POSITIVE LOGITS
     inherently
    0.68
     traditionally
    0.59
     intrinsically
    0.57
     generally
    0.55
     broadly
    0.55
     primarily
    0.55
     increasingly
    0.54
     fundamentally
    0.54
     pervasive
    0.52
    MODE
    0.52
    Activations Density 21.402%

    No Known Activations