© 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. 20435
    Prev
    Next
    INDEX
    Explanations

    statements indicating action or reaction in response to a specific situation or event

    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
    709
    +0.12
    0.6%
    706
    +0.11
    0.5%
    439
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    112
    +0.12
    0.02
    439
    +0.11
    0.02
    706
    +0.11
    0.02
    Negative Logits
     GEAR
    -0.79
    ãģĨ
    -0.72
     à¨
    -0.67
     Pebble
    -0.65
     Anchorage
    -0.61
     handwriting
    -0.60
     buckle
    -0.60
     Bord
    -0.60
     misunder
    -0.59
     sheet
    -0.58
    POSITIVE LOGITS
    becca
    1.31
    levant
    1.28
    asons
    1.23
    venge
    1.19
    asonable
    1.18
    venue
    1.14
    ggie
    1.14
    ceived
    1.13
    ceiver
    1.12
    verend
    1.11
    Activations Density 0.012%

    No Known Activations