© 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. 8166
    Prev
    Next
    INDEX
    Explanations

    phrases related to explosive actions or events

    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
    288
    +0.13
    0.6%
    526
    +0.11
    0.5%
    203
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    531
    +0.13
    0.02
    346
    +0.11
    0.02
    288
    +0.10
    0.02
    Negative Logits
    laus
    -0.81
     Princ
    -0.69
    éĸ
    -0.67
    DAY
    -0.65
    dit
    -0.65
    SEA
    -0.64
     Legends
    -0.63
     Inquis
    -0.62
     Virt
    -0.62
    ç¥ŀ
    -0.62
    POSITIVE LOGITS
    arding
    1.33
    arded
    1.30
    astically
    1.06
    astic
    1.03
    ards
    1.01
     detonated
    0.91
     deton
    0.91
    ard
    0.90
    blast
    0.89
    proof
    0.87
    Activations Density 0.024%

    No Known Activations