© 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
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 49152
    5. 8-RES_FS49152-JB
    6. 41486
    Prev
    Next
    INDEX
    Explanations

    references to specific brand names related to firearms

    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
    630
    +0.10
    0.4%
    497
    +0.10
    0.4%
    90
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    406
    +0.10
    0.02
    659
    +0.10
    0.02
    299
    +0.10
    0.01
    Negative Logits
    catentry
    -0.71
    PLIED
    -0.66
    ionage
    -0.65
    ascript
    -0.65
    glers
    -0.64
    ruary
    -0.62
     MER
    -0.60
     Revenge
    -0.59
    士
    -0.58
     trope
    -0.58
    POSITIVE LOGITS
    nikov
    0.97
    chuk
    0.94
    akra
    0.86
    ubi
    0.85
    din
    0.83
    gaard
    0.83
    nick
    0.83
    sky
    0.82
    odan
    0.81
    inski
    0.79
    Activations Density 0.020%

    No Known Activations