© 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. 9833
    Prev
    Next
    INDEX
    Explanations

    text related to software or product features and functionalities

    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
    679
    +0.14
    0.6%
    96
    +0.12
    0.5%
    575
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    679
    +0.14
    0.04
    540
    +0.12
    0.04
    575
    +0.11
    0.03
    Negative Logits
    kus
    -0.92
    oner
    -0.76
    yss
    -0.75
    inion
    -0.71
    luck
    -0.70
    prus
    -0.69
    sych
    -0.67
    lat
    -0.64
    irds
    -0.63
    lyak
    -0.62
    POSITIVE LOGITS
     booklet
    0.96
     text
    0.84
    TextColor
    0.77
     manuals
    0.76
     texts
    0.75
     spelled
    0.75
     provided
    0.74
    ript
    0.73
     description
    0.73
     supplied
    0.72
    Activations Density 0.148%

    No Known Activations