© 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. 8022
    Prev
    Next
    INDEX
    Explanations

    words related to morality and ethical behavior

    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
    746
    +0.11
    0.5%
    89
    +0.10
    0.5%
    515
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    746
    +0.11
    0.01
    515
    +0.10
    0.01
    406
    +0.09
    0.01
    Negative Logits
    ierrez
    -0.65
    izoph
    -0.60
     coasts
    -0.60
    ritch
    -0.59
    rooms
    -0.57
     awhile
    -0.56
     4090
    -0.56
     cooldown
    -0.55
    redo
    -0.54
     Estimated
    -0.54
    POSITIVE LOGITS
    iak
    0.94
    stru
    0.89
    line
    0.85
    bol
    0.85
    lines
    0.84
    ko
    0.83
    ais
    0.83
    ī
    0.80
    SHIP
    0.79
    opter
    0.78
    Activations Density 0.029%

    No Known Activations