© 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. 4902
    Prev
    Next
    INDEX
    Explanations

    phrases or sentences containing the word "whoever"

    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
    +0.15
    0.8%
    735
    +0.09
    0.5%
    718
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    393
    +0.15
    0.25
    362
    +0.09
    0.11
    526
    +0.09
    -0.08
    Negative Logits
    enthal
    -0.80
    eda
    -0.74
    eed
    -0.62
    chell
    -0.62
    lander
    -0.62
    ey
    -0.62
    ert
    -0.61
    aci
    -0.60
    iven
    -0.60
    press
    -0.58
    POSITIVE LOGITS
    soever
    0.96
     else
    0.84
    omever
    0.76
    ftime
    0.74
     desired
    0.69
     interfered
    0.67
     betrayed
    0.66
     disliked
    0.65
     offended
    0.64
     whoever
    0.63
    Activations Density 10.041%

    No Known Activations