© 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 E2E Recon
    9. 6-RES_SCEFR-AJT
    11. 8550
    Prev
    Next
    INDEX
    Explanations

    phrases related to childhood experiences, such as trauma, schooling, dreams, and playground

    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
    288
    +0.24
    0.8%
    99
    +0.17
    0.6%
    575
    +0.17
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.24
    0.12
    575
    +0.17
    0.11
    132
    +0.17
    0.08
    Negative Logits
    abase
    -1.24
    andum
    -1.22
    chens
    -1.04
    abases
    -1.04
    utical
    -1.00
    aws
    -0.98
    ioxide
    -0.98
    aeda
    -0.97
    oute
    -0.96
    oples
    -0.93
    POSITIVE LOGITS
     è£ıè
    0.92
    ishly
    0.91
    scape
    0.89
     adversity
    0.85
     innocence
    0.84
     experien
    0.82
     memories
    0.82
     indisc
    0.81
     Pradesh
    0.81
    s
    0.80
    Activations Density 1.277%

    No Known Activations