Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 3072
    5. 8-RES_FS3072-JB
    6. 2415
    Prev
    Next
    INDEX
    Explanations

    references to minors or young individuals, especially in contexts related to sexual misconduct or abuse

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    433
    +0.11
    0.5%
    35
    +0.11
    0.5%
    274
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    302
    +0.11
    0.08
    433
    +0.11
    0.07
    383
    +0.10
    0.07
    Negative Logits
    ichick
    -0.76
    nce
    -0.72
    lyak
    -0.71
    neum
    -0.70
    anyahu
    -0.68
    icist
    -0.68
    rosis
    -0.68
    rov
    -0.67
    leased
    -0.67
    orius
    -0.66
    POSITIVE LOGITS
     boys
    1.21
     girls
    1.12
     Scouts
    1.08
     Boys
    1.02
    boys
    1.00
     kids
    0.94
     Girls
    0.92
    kids
    0.92
     boy
    0.92
    Girls
    0.90
    Activations Density 0.672%

    No Known Activations