© 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 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 E2E Recon
    9. 6-RES_SCEFR-AJT
    11. 956
    Prev
    Next
    INDEX
    Explanations

    phrases related to a person named Len

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scefr-ajt/6-res_scefr-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
    285
    +0.21
    0.8%
    356
    +0.14
    0.5%
    114
    +0.13
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.21
    0.06
    288
    +0.14
    0.07
    285
    +0.13
    0.03
    Negative Logits
     Camel
    -0.88
     Virgin
    -0.78
    fly
    -0.78
     Debor
    -0.77
    tl
    -0.77
     Presidency
    -0.76
    AGES
    -0.75
     Occupations
    -0.74
     Leap
    -0.73
    MODE
    -0.73
    POSITIVE LOGITS
    ovo
    1.28
    iency
    1.24
    essee
    1.15
    ately
    1.06
    ient
    1.04
    ikuman
    1.02
    apse
    0.99
    ishers
    0.97
    opoly
    0.97
    heed
    0.96
    Activations Density 0.833%

    No Known Activations