© 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 12288
    9. 8-RES_FS12288-JB
    11. 9774
    Prev
    Next
    INDEX
    Explanations

    words related to specific names or terms, particularly those with unusual capitalization patterns

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    172
    +0.13
    0.6%
    107
    +0.11
    0.5%
    364
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    107
    +0.13
    0.03
    701
    +0.11
    0.03
    364
    +0.10
    0.03
    Negative Logits
    aturdays
    -0.69
    ufact
    -0.66
    PDATE
    -0.65
    steen
    -0.64
    heit
    -0.63
    FML
    -0.62
    BG
    -0.62
    leneck
    -0.62
    bender
    -0.62
    brance
    -0.61
    POSITIVE LOGITS
    reme
    0.90
    Large
    0.80
    ©¶æ
    0.74
    abytes
    0.71
    xes
    0.71
    thur
    0.69
    modules
    0.68
    509
    0.67
    toc
    0.67
    uni
    0.67
    Activations Density 0.303%

    No Known Activations