© 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. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 12288
    9. 8-RES_FS12288-JB
    11. 10648
    Prev
    Next
    INDEX
    Explanations

    phrases related to emotional attachment or holding on to something

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    Data Type
    torch.float32
    Hook Point
    blocks.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    373
    +0.13
    0.6%
    273
    +0.10
    0.4%
    526
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    269
    +0.13
    0.03
    90
    +0.10
    0.03
    625
    +0.10
    0.03
    Negative Logits
    cpp
    -0.71
    Ķ
    -0.69
    Ĥª
    -0.67
    çĶ°
    -0.67
    enegger
    -0.66
    £
    -0.64
    ©
    -0.64
    °
    -0.63
    BUT
    -0.62
    ¦
    -0.61
    POSITIVE LOGITS
     behalf
    1.18
    shore
    1.01
    erous
    0.98
    eday
    0.87
    coming
    0.84
    unal
    0.81
    side
    0.78
    eness
    0.77
     board
    0.73
     toes
    0.71
    Activations Density 0.148%

    No Known Activations