© 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 24576
    9. 8-RES_FS24576-JB
    11. 7321
    Prev
    Next
    INDEX
    Explanations

    mentions of arrests

    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
    447
    +0.18
    0.8%
    665
    +0.13
    0.6%
    380
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    665
    +0.18
    0.04
    380
    +0.13
    0.03
    630
    +0.10
    0.03
    Negative Logits
     Remastered
    -0.72
    ankind
    -0.70
    issance
    -0.69
    dust
    -0.69
    rium
    -0.68
    emic
    -0.67
    casters
    -0.65
    clips
    -0.64
     superiority
    -0.63
    layout
    -0.61
    POSITIVE LOGITS
     criminally
    0.91
     felon
    0.81
     handcuffed
    0.80
     arrested
    0.74
     illegally
    0.74
     alongside
    0.72
     unlawfully
    0.72
     administr
    0.72
     arra
    0.71
    untarily
    0.71
    Activations Density 0.059%

    No Known Activations