© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    3. GPT2-Small
    4. Residual Similar CE E2E Recon
    5. 6-RES_SCEFR-AJT
    6. 1899
    Prev
    Next
    INDEX
    Explanations

    references to a specific person named Luke, possibly related to sports or media

    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
    326
    +0.23
    0.8%
    527
    +0.17
    0.6%
    26
    +0.16
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.23
    0.07
    659
    +0.17
    0.06
    288
    +0.16
    0.07
    Negative Logits
    ãĤ¢ãĥ«
    -1.30
    ursed
    -1.23
     reluct
    -1.17
    ktop
    -1.15
    gency
    -1.09
     Seym
    -1.03
     pregn
    -1.00
    ãĤ½
    -0.99
    Ùħ
    -0.99
     Parenthood
    -0.96
    POSITIVE LOGITS
    warm
    2.07
     Skywalker
    1.95
     Cage
    1.61
    ership
    1.20
    lihood
    1.18
     Vader
    1.11
     Gat
    1.07
    avin
    1.07
     Luke
    1.05
    antine
    1.04
    Activations Density 0.926%

    No Known Activations