Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 1536
    5. 8-RES_FS1536-JB
    6. 508
    Prev
    Next
    INDEX
    Explanations

    locations or distances expressed in relation to a reference point

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    180
    +0.09
    0.4%
    464
    +0.09
    0.4%
    323
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    669
    +0.09
    0.08
    608
    +0.09
    0.10
    362
    +0.09
    0.09
    Negative Logits
    OY
    -0.79
    CW
    -0.64
    PDF
    -0.63
    zhen
    -0.63
    Anonymous
    -0.62
    nell
    -0.60
    Sort
    -0.60
    Includes
    -0.60
    raq
    -0.60
    DL
    -0.60
    POSITIVE LOGITS
     midnight
    0.86
     shore
    0.85
     Interstate
    0.83
     sunset
    0.82
     sunrise
    0.78
     campus
    0.77
    rium
    0.77
     landfall
    0.76
    sighted
    0.76
     horizon
    0.76
    Activations Density 2.014%

    No Known Activations