© 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. 24507
    Prev
    Next
    INDEX
    Explanations

    terms related to tilting or leaning, often metaphorically to indicate bias or preference

    oai_token-act-pair · gpt-4o-miniTriggered 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
    538
    +0.11
    0.5%
    278
    +0.10
    0.4%
    349
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    665
    +0.11
    0.04
    195
    +0.10
    0.02
    125
    +0.09
    0.02
    Negative Logits
    ufact
    -0.77
     Liberation
    -0.70
    zek
    -0.70
     Mub
    -0.69
     Plain
    -0.67
    ============
    -0.65
    riot
    -0.65
    ilk
    -0.64
    alls
    -0.64
    COMPLE
    -0.63
    POSITIVE LOGITS
     toward
    1.49
     towards
    1.42
     downwards
    1.24
     downward
    1.22
     upward
    1.12
     favour
    1.08
     Towards
    1.05
     favoring
    1.03
    oward
    1.02
     upwards
    0.99
    Activations Density 0.126%

    No Known Activations