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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 46002
    Prev
    Next
    INDEX
    Explanations

    names of people, particularly the name "Peterson"

    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
    351
    +0.17
    0.8%
    464
    +0.13
    0.6%
    447
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.17
    0.26
    326
    +0.13
    0.22
    464
    +0.11
    0.11
    Negative Logits
    creen
    -0.89
    UV
    -0.81
    rooms
    -0.78
    hazard
    -0.77
    ume
    -0.75
    uten
    -0.74
     negotiator
    -0.73
    wave
    -0.71
    chery
    -0.69
    ipeg
    -0.69
    POSITIVE LOGITS
     Peterson
    0.83
    asus
    0.80
    mented
    0.78
    lund
    0.74
    heses
    0.74
     Ake
    0.70
     Rivals
    0.68
    nered
    0.64
    âĶĢ
    0.64
    ando
    0.63
    Activations Density 6.874%

    No Known Activations