© 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. 12656
    Prev
    Next
    INDEX
    Explanations

    references related to the Obama administration and political figures such as Obama and Romney

    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
    527
    +0.16
    0.5%
    343
    +0.15
    0.5%
    237
    +0.14
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    346
    +0.16
    0.06
    254
    +0.15
    0.06
    15
    +0.14
    0.07
    Negative Logits
     Cth
    -1.09
     Ukrain
    -1.08
    terday
    -1.08
    ategory
    -1.07
    ongh
    -1.05
     Flavoring
    -1.03
    veyard
    -0.97
    asma
    -0.96
     Azerb
    -0.95
    TOR
    -0.94
    POSITIVE LOGITS
    Care
    1.23
     administration
    1.18
    sonian
    1.08
    Obama
    0.97
     Obama
    0.97
    ovic
    0.96
     advisor
    0.95
     Administration
    0.95
    shire
    0.95
    ufact
    0.95
    Activations Density 0.918%

    No Known Activations