© 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. 2495
    Prev
    Next
    INDEX
    Explanations

    references to the word "Rose"

    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
    166
    +0.12
    0.5%
    571
    +0.12
    0.4%
    504
    +0.12
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.12
    0.08
    659
    +0.12
    0.06
    288
    +0.12
    0.09
    Negative Logits
    redit
    -1.09
    ulhu
    -1.02
    SPONSORED
    -0.99
    etheless
    -0.95
    essions
    -0.91
    ernandez
    -0.89
    eatures
    -0.86
    adden
    -0.86
    orsi
    -0.86
     nomine
    -0.86
    POSITIVE LOGITS
    mary
    1.98
    anne
    1.44
    xual
    1.29
    velt
    1.26
    hill
    1.25
    mont
    1.24
     Quartz
    1.16
    Rose
    1.16
    dale
    1.13
    water
    1.12
    Activations Density 1.307%

    No Known Activations