© 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 24576
    9. 8-RES_FS24576-JB
    11. 8029
    Prev
    Next
    INDEX
    Explanations

    pronouns and names of 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_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    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
    447
    +0.26
    1.3%
    480
    +0.26
    1.3%
    481
    +0.13
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    139
    +0.26
    0.07
    492
    +0.26
    0.05
    762
    +0.13
    0.07
    Negative Logits
     Seym
    -0.79
     pregn
    -0.60
     Leban
    -0.60
    classic
    -0.59
    etheless
    -0.59
     Composite
    -0.56
    itaire
    -0.56
    igmatic
    -0.56
     +/-
    -0.54
    Operation
    -0.54
    POSITIVE LOGITS
    zbollah
    1.20
    resy
    0.95
     Majesty
    0.90
    ading
    0.83
    gemony
    0.83
    'll
    0.83
    mos
    0.82
    eded
    0.81
    eding
    0.81
    brew
    0.81
    Activations Density 0.206%

    No Known Activations