© 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. 2790
    Prev
    Next
    INDEX
    Explanations

    locations and organizations related to conflict and warfare

    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
    486
    +0.10
    0.5%
    759
    +0.09
    0.4%
    513
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    284
    +0.10
    0.05
    659
    +0.09
    0.05
    232
    +0.09
    0.04
    Negative Logits
     BAD
    -0.71
    angelo
    -0.65
     Helpful
    -0.65
    ortment
    -0.65
    erella
    -0.59
    eteria
    -0.58
     describ
    -0.58
    inctions
    -0.57
    auga
    -0.57
    osph
    -0.56
    POSITIVE LOGITS
    schild
    0.81
    isphere
    0.70
    åĤ
    0.68
    illance
    0.67
    ministic
    0.63
    plugin
    0.62
    addons
    0.62
    oof
    0.61
    arian
    0.61
    ĪĴ
    0.60
    Activations Density 0.254%

    No Known Activations