© 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. 12515
    Prev
    Next
    INDEX
    Explanations

    long ellipses indicating a break in text or thought

    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
    559
    +0.07
    0.5%
    481
    +0.06
    0.5%
    86
    +0.06
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    420
    +0.07
    0.01
    38
    +0.06
    0.01
    607
    +0.06
    0.01
    Negative Logits
    kinson
    -0.48
    itialized
    -0.42
    ĪĴ
    -0.42
    itton
    -0.39
     princ
    -0.36
     misunder
    -0.36
     livest
    -0.35
    uristic
    -0.34
    kson
    -0.34
    itarian
    -0.34
    POSITIVE LOGITS
    aple
    0.37
    ys
    0.34
     Trout
    0.33
    Tax
    0.33
     ICE
    0.33
     Shows
    0.32
    merce
    0.31
    ynthesis
    0.31
     Robin
    0.31
    xy
    0.31
    Activations Density 0.002%

    No Known Activations