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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 24576
    5. 8-RES_FS24576-JB
    6. 385
    Prev
    Next
    INDEX
    Explanations

    phrases related to the exchange of information and experiences among individuals and communities

    oai_token-act-pair · gpt-3.5-turboTriggered 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
    638
    +0.11
    0.5%
    709
    +0.10
    0.5%
    84
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    638
    +0.11
    0.03
    161
    +0.10
    0.03
    155
    +0.10
    0.03
    Negative Logits
     teasp
    -0.96
    OGR
    -0.75
    ©¶æ¥µ
    -0.74
    oland
    -0.71
    photo
    -0.70
    aunder
    -0.69
    quez
    -0.68
    à¥
    -0.68
    cue
    -0.66
    ा
    -0.66
    POSITIVE LOGITS
     how
    1.13
     why
    0.88
     ourselves
    0.85
     whether
    0.82
     what
    0.78
     regards
    0.75
     them
    0.75
     him
    0.73
     our
    0.72
     calib
    0.72
    Activations Density 0.096%

    No Known Activations