© 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
    3. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    5. GPT2-Small
    7. Residual Similar CE Local
    9. 6-RES_SCL-AJT
    11. 39592
    Prev
    Next
    INDEX
    Explanations

    proper nouns related to the name "Rose"

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scl-ajt/6-res_scl-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
    447
    +0.20
    0.9%
    284
    +0.12
    0.5%
    571
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    543
    +0.20
    0.08
    343
    +0.12
    0.07
    326
    +0.09
    0.22
    Negative Logits
    ulhu
    -0.77
    cffff
    -0.71
    essions
    -0.71
    etheless
    -0.69
    redit
    -0.69
    anooga
    -0.67
    ernandez
    -0.66
    rongh
    -0.65
    SPONSORED
    -0.65
    ravel
    -0.63
    POSITIVE LOGITS
    mary
    1.58
    anne
    1.06
    mont
    1.01
    velt
    0.99
    xual
    0.96
    hill
    0.96
    water
    0.96
     Quartz
    0.96
    Rose
    0.94
    mond
    0.93
    Activations Density 9.618%

    No Known Activations