© 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 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. 12357
    Prev
    Next
    INDEX
    Explanations

    violent and aggressive actions or dialogues

    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
    419
    +0.13
    0.6%
    371
    +0.10
    0.5%
    282
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    678
    +0.13
    0.05
    371
    +0.10
    0.04
    419
    +0.10
    0.04
    Negative Logits
    ontent
    -0.82
     specialization
    -0.80
     unemploy
    -0.75
    igon
    -0.74
    Sov
    -0.72
     Colomb
    -0.71
     redundancy
    -0.69
     optimization
    -0.68
    CLUS
    -0.68
     specialize
    -0.67
    POSITIVE LOGITS
    stretched
    1.29
     gently
    1.27
     clasp
    1.14
     grasped
    1.12
     fingers
    1.08
     clenched
    1.06
     knees
    1.06
     lips
    1.05
     palms
    1.04
     palm
    1.03
    Activations Density 0.315%

    No Known Activations