© 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. 1717
    Prev
    Next
    INDEX
    Explanations

    the word 'Die', particularly in the context of negative events or situations

    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
    98
    +0.11
    0.5%
    8
    +0.10
    0.5%
    383
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    759
    +0.11
    0.03
    319
    +0.10
    0.03
    223
    +0.10
    0.03
    Negative Logits
    arov
    -0.76
    ajo
    -0.71
    orney
    -0.69
    Īè
    -0.67
    ropri
    -0.67
    aired
    -0.66
     Ally
    -0.64
    abulary
    -0.62
     propri
    -0.59
    afort
    -0.58
    POSITIVE LOGITS
    hard
    1.22
    getic
    1.05
    ffen
    1.03
    lect
    0.90
    bold
    0.83
     reckoning
    0.83
     horribly
    0.79
    hl
    0.76
     miser
    0.73
    hyde
    0.72
    Activations Density 0.084%

    No Known Activations