© 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 49152
    9. 8-RES_FS49152-JB
    11. 25712
    Prev
    Next
    INDEX
    Explanations

    references to rockets and missile technology

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    +0.17
    0.8%
    269
    +0.13
    0.6%
    268
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    269
    +0.17
    0.01
    288
    +0.13
    0.01
    286
    +0.10
    0.01
    Negative Logits
    redo
    -0.88
     Borders
    -0.87
    tale
    -0.79
    DonaldTrump
    -0.75
    女
    -0.71
    "]=>
    -0.67
    ktop
    -0.67
    SPONSORED
    -0.65
     Smile
    -0.64
    代
    -0.64
    POSITIVE LOGITS
     propulsion
    1.06
    eers
    1.04
     launcher
    1.02
     propelled
    1.01
    ulic
    0.99
     launchers
    0.97
    laun
    0.96
     propell
    0.95
     motors
    0.94
    eer
    0.93
    Activations Density 0.012%

    No Known Activations