© 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. 10703
    Prev
    Next
    INDEX
    Explanations

    terms related to negative emotions and situations, particularly in a political or legal context

    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
    611
    +0.11
    0.5%
    218
    +0.10
    0.5%
    640
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    611
    +0.11
    0.06
    362
    +0.10
    0.06
    476
    +0.09
    0.04
    Negative Logits
    HK
    -0.80
    amount
    -0.77
    quel
    -0.74
    ily
    -0.73
    ãĥĹ
    -0.71
    arella
    -0.69
    rang
    -0.69
    hester
    -0.68
    CRIPTION
    -0.67
    ives
    -0.66
    POSITIVE LOGITS
     civilisation
    0.79
     civilization
    0.76
     unsuspecting
    0.75
     legality
    0.74
     Appalach
    0.68
     academia
    0.68
     fame
    0.67
     darkness
    0.65
     empires
    0.65
    rium
    0.64
    Activations Density 0.347%

    No Known Activations