© 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. 24387
    Prev
    Next
    INDEX
    Explanations

    phrases indicating a necessity or demand for assistance or resources

    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
    762
    +0.12
    0.5%
    16
    +0.10
    0.5%
    581
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    762
    +0.12
    0.02
    125
    +0.10
    0.02
    16
    +0.10
    0.02
    Negative Logits
    irted
    -0.62
     Buckley
    -0.60
    alter
    -0.59
    ells
    -0.58
    authorized
    -0.56
     AVG
    -0.56
     Ging
    -0.55
     Mou
    -0.55
     Init
    -0.54
     avg
    -0.54
    POSITIVE LOGITS
    Ø©
    0.89
    iatric
    0.78
    angering
    0.71
    liction
    0.69
     financially
    0.68
    ï¸
    0.67
    FUL
    0.66
    OGR
    0.66
    ful
    0.65
    nered
    0.65
    Activations Density 0.051%

    No Known Activations