© 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. 2153
    Prev
    Next
    INDEX
    Explanations

    long-term economic and political discussions, especially related to government policies and societal impacts

    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
    706
    +0.17
    0.6%
    288
    +0.14
    0.5%
    715
    +0.13
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    666
    +0.17
    0.04
    231
    +0.14
    0.04
    706
    +0.13
    0.04
    Negative Logits
     Cosponsors
    -1.57
    rote
    -1.18
    hou
    -1.16
    HOU
    -1.09
    chu
    -1.05
     DRAG
    -1.03
    yp
    -1.02
    gyn
    -1.01
     ItemLevel
    -0.97
    illard
    -0.97
    POSITIVE LOGITS
     stability
    1.21
     viability
    1.20
     commitment
    1.06
     memory
    1.03
     unemployed
    1.03
     upkeep
    1.01
     lasting
    1.01
     outcomes
    1.00
     enjoyment
    1.00
     gratification
    0.99
    Activations Density 0.669%

    No Known Activations