© 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. 7534
    Prev
    Next
    INDEX
    Explanations

    motivations and actions related to indulgence, potentially in a negative or excessive manner

    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
    624
    +0.13
    0.6%
    301
    +0.13
    0.6%
    541
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    746
    +0.13
    0.02
    624
    +0.13
    0.02
    288
    +0.12
    0.03
    Negative Logits
    è£ħ
    -0.81
    prototype
    -0.74
    ĺħ
    -0.73
     Shack
    -0.69
    ACP
    -0.69
    Hop
    -0.68
    PLA
    -0.68
    hair
    -0.68
    OHN
    -0.67
    \\\\\\\\
    -0.66
    POSITIVE LOGITS
    atory
    1.35
    gence
    1.18
    gments
    1.10
    ation
    1.07
    iment
    1.06
    icates
    1.00
    iments
    1.00
    ication
    1.00
    ices
    1.00
    ences
    1.00
    Activations Density 0.057%

    No Known Activations