© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 16279
    Prev
    Next
    INDEX
    Explanations

    phrases related to stories or historical events

    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
    362
    +0.14
    0.7%
    18
    +0.10
    0.5%
    367
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.14
    0.06
    614
    +0.10
    0.04
    620
    +0.10
    0.03
    Negative Logits
    女
    -0.81
    é¾
    -0.69
    909
    -0.67
    à©
    -0.65
    ril
    -0.64
     âľ
    -0.64
    в
    -0.63
    verages
    -0.62
    KK
    -0.61
    ername
    -0.60
    POSITIVE LOGITS
    geries
    1.28
     sale
    1.08
    bidden
    1.06
    gery
    1.05
     purposes
    0.93
     themselves
    0.87
    ked
    0.85
     reuse
    0.84
     ourselves
    0.83
     hire
    0.82
    Activations Density 0.170%

    No Known Activations