© 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
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 24576
    5. 8-RES_FS24576-JB
    6. 18820
    Prev
    Next
    INDEX
    Explanations

    specific named entities or entities related to contact, communication, and actions like dropping or arresting

    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
    366
    +0.12
    0.5%
    404
    +0.12
    0.5%
    739
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    636
    +0.12
    0.02
    366
    +0.12
    0.02
    298
    +0.09
    0.02
    Negative Logits
    âĸ¬
    -0.72
    MAL
    -0.71
    WM
    -0.69
    CLA
    -0.68
    SELECT
    -0.67
    APH
    -0.67
    SEA
    -0.66
    WR
    -0.64
    ä¸ī
    -0.64
    orth
    -0.64
    POSITIVE LOGITS
     bombshell
    1.06
     bombs
    0.94
     leaflets
    0.89
     hints
    0.88
     towel
    0.79
     curtain
    0.79
     inhib
    0.78
     stairs
    0.78
     hammer
    0.76
     balloons
    0.76
    Activations Density 0.087%

    No Known Activations