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

    Neuronpedia

    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. 1885
    Prev
    Next
    INDEX
    Explanations

    references to legal and political concepts

    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
    425
    +0.12
    0.5%
    690
    +0.10
    0.5%
    563
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    690
    +0.12
    0.02
    425
    +0.10
    0.02
    97
    +0.10
    0.02
    Negative Logits
     ILCS
    -0.76
    asks
    -0.71
    adas
    -0.70
    auga
    -0.70
    needed
    -0.68
    >:
    -0.67
     flashbacks
    -0.67
    idays
    -0.66
    idel
    -0.61
    alan
    -0.61
    POSITIVE LOGITS
     thumb
    0.94
     therm
    0.87
     propri
    0.87
     physics
    0.85
     prud
    0.77
     precedence
    0.77
    clud
    0.73
     Conduct
    0.72
     decency
    0.70
     conduct
    0.69
    Activations Density 0.084%

    No Known Activations