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 49152
    5. 8-RES_FS49152-JB
    6. 40070
    Prev
    Next
    INDEX
    Explanations

    special characters and formatting symbols

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    314
    +0.22
    1.0%
    160
    +0.14
    0.6%
    102
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    314
    +0.22
    0.01
    160
    +0.14
    0.02
    417
    +0.12
    0.01
    Negative Logits
     seiz
    -0.76
     aid
    -0.73
     lod
    -0.72
     Mechdragon
    -0.72
     Seym
    -0.72
     guid
    -0.70
     nodd
    -0.69
     Swordsman
    -0.69
     Navajo
    -0.67
     Niet
    -0.67
    POSITIVE LOGITS
    insert
    1.28
    blank
    1.25
    yet
    1.15
    REDACTED
    1.12
    #$
    1.12
    fac
    1.08
    never
    1.06
    anything
    1.06
    their
    1.05
    future
    1.03
    Activations Density 0.024%

    No Known Activations