© 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. 10533
    Prev
    Next
    INDEX
    Explanations

    mentions of retirement or retired individuals

    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
    288
    +0.14
    0.7%
    27
    +0.09
    0.4%
    191
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.14
    0.03
    100
    +0.09
    0.02
    571
    +0.09
    0.02
    Negative Logits
    abal
    -0.80
    iola
    -0.79
    owitz
    -0.75
    otin
    -0.67
     Grimm
    -0.64
    eta
    -0.64
    matic
    -0.63
    insk
    -0.63
    ñ
    -0.61
    ythm
    -0.60
    POSITIVE LOGITS
     retiring
    0.92
     retire
    0.87
     pensions
    0.81
     retirement
    0.78
     obsolete
    0.77
    spective
    0.77
    dinand
    0.76
    irements
    0.71
     redundant
    0.71
     veter
    0.70
    Activations Density 0.079%

    No Known Activations