© 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 3072
    9. 8-RES_FS3072-JB
    11. 3006
    Prev
    Next
    INDEX
    Explanations

    actions and scenarios related to making a physical change or adjustment

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    90
    +0.10
    0.5%
    761
    +0.10
    0.4%
    92
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    761
    +0.10
    0.11
    184
    +0.10
    0.06
    665
    +0.10
    0.10
    Negative Logits
    ère
    -0.87
    si
    -0.68
    eria
    -0.65
     covering
    -0.65
    nings
    -0.64
     requesting
    -0.64
    iere
    -0.64
    tes
    -0.64
    word
    -0.64
     barring
    -0.63
    POSITIVE LOGITS
     oneself
    1.27
     Yourself
    0.89
    redients
    0.88
     yourself
    0.83
    HAM
    0.72
     someone
    0.69
     engagements
    0.68
    BuyableInstoreAndOnline
    0.67
     ourselves
    0.67
     Pengu
    0.65
    Activations Density 1.458%

    No Known Activations