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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 22883
    Prev
    Next
    INDEX
    Explanations

    references to military actions and casualties in conflict zones

    oai_token-act-pair · gpt-4o-miniTriggered 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
    745
    +0.11
    0.5%
    180
    +0.11
    0.5%
    547
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    547
    +0.11
    0.04
    180
    +0.11
    0.05
    412
    +0.11
    0.04
    Negative Logits
    Consumer
    -0.87
    wcsstore
    -0.84
     chirop
    -0.82
    aughs
    -0.82
    olson
    -0.81
    Scot
    -0.81
    URA
    -0.76
     USPS
    -0.76
    Russell
    -0.73
    XT
    -0.72
    POSITIVE LOGITS
     outskirts
    1.16
     Aleppo
    1.15
     Mosul
    1.15
     Idlib
    1.14
     Donetsk
    1.13
     Raqqa
    1.12
     besieged
    1.10
     Homs
    1.08
     Syria
    1.06
     siege
    1.03
    Activations Density 0.361%

    No Known Activations