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

    Neuronpedia

    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. 24026
    Prev
    Next
    INDEX
    Explanations

    phrases indicating levels of attention or perceived importance

    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
    408
    +0.13
    0.6%
    699
    +0.10
    0.5%
    279
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    408
    +0.13
    0.07
    279
    +0.10
    0.06
    13
    +0.10
    0.06
    Negative Logits
    ²¾
    -0.73
    ħĭ
    -0.70
    anwhile
    -0.68
    raid
    -0.68
    nown
    -0.65
    uggest
    -0.65
    ctors
    -0.65
    ocene
    -0.63
    artments
    -0.63
    ramid
    -0.63
    POSITIVE LOGITS
     firepower
    1.25
     notoriety
    1.25
     traction
    1.23
     attention
    1.16
     clout
    1.15
     fame
    1.08
     publicity
    1.06
     insight
    1.01
     momentum
    1.00
     autonomy
    0.99
    Activations Density 0.243%

    No Known Activations