© 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. 22353
    Prev
    Next
    INDEX
    Explanations

    the concept of determination or resolve in various contexts

    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
    111
    +0.11
    0.5%
    296
    +0.10
    0.5%
    624
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    111
    +0.11
    0.01
    624
    +0.10
    0.01
    417
    +0.10
    0.01
    Negative Logits
    atha
    -0.83
    ILA
    -0.81
    vertisement
    -0.76
    ocular
    -0.72
    umbn
    -0.71
    pheus
    -0.69
    acl
    -0.69
    rompt
    -0.69
    essage
    -0.68
    PLIED
    -0.68
    POSITIVE LOGITS
    lessly
    0.99
    lessness
    0.88
    ly
    0.88
    LY
    0.78
    ially
    0.76
     determination
    0.71
    less
    0.68
    bred
    0.68
    staking
    0.66
    ively
    0.62
    Activations Density 0.027%

    No Known Activations