© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 6116
    Prev
    Next
    INDEX
    Explanations

    instances of aggression or conflict

    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
    28
    +0.12
    0.5%
    374
    +0.11
    0.5%
    321
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    608
    +0.12
    0.07
    686
    +0.11
    0.04
    283
    +0.09
    0.02
    Negative Logits
    ãĥīãĥ©ãĤ´ãĥ³
    -0.70
    ãĥĢ
    -0.70
    tera
    -0.69
     sidx
    -0.67
    pic
    -0.66
    pir
    -0.65
    ãĥĩãĤ£
    -0.63
    Fall
    -0.63
    aspberry
    -0.62
    iac
    -0.61
    POSITIVE LOGITS
     authorities
    0.95
     police
    0.87
     themselves
    0.82
     strangers
    0.79
     administrators
    0.76
     him
    0.76
     policemen
    0.75
     supervisors
    0.74
     landlords
    0.74
     careg
    0.70
    Activations Density 0.756%

    No Known Activations