© 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. 41812
    Prev
    Next
    INDEX
    Explanations

    the phrase "ha" as a recurring element in various contexts

    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
    612
    +0.10
    0.4%
    372
    +0.08
    0.4%
    343
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.10
    0.01
    201
    +0.08
    0.01
    502
    +0.08
    0.01
    Negative Logits
    atories
    -0.75
     outgoing
    -0.68
    inarily
    -0.65
    é»Ĵ
    -0.64
    rations
    -0.62
    night
    -0.62
    bats
    -0.60
     Collider
    -0.60
    rored
    -0.60
    ocity
    -0.59
    POSITIVE LOGITS
    wn
    1.26
    user
    1.18
    illard
    1.04
    pta
    0.97
    fter
    0.95
    emi
    0.92
    ppa
    0.92
    irst
    0.89
    ñ
    0.89
    pp
    0.89
    Activations Density 0.008%

    No Known Activations