© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 49152
    9. 8-RES_FS49152-JB
    11. 37331
    Prev
    Next
    INDEX
    Explanations

    names of individuals associated with significant events or actions

    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
    481
    +0.21
    1.0%
    447
    +0.13
    0.6%
    141
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.21
    0.07
    232
    +0.13
    0.05
    700
    +0.12
    0.04
    Negative Logits
    LEASE
    -0.70
    anwhile
    -0.70
    avascript
    -0.63
    regor
    -0.61
     thrilling
    -0.61
    ensional
    -0.61
     confir
    -0.61
    acters
    -0.60
     hinge
    -0.60
     collisions
    -0.59
    POSITIVE LOGITS
    *,
    0.95
     Jr
    0.84
     Sr
    0.84
    ,
    0.81
     QC
    0.75
     III
    0.73
     Pres
    0.70
    !,
    0.67
     Gonzalez
    0.67
    ?,
    0.65
    Activations Density 0.208%

    No Known Activations