© 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. 45870
    Prev
    Next
    INDEX
    Explanations

    references to mascots and characters in a humorous context

    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
    447
    +0.28
    1.3%
    480
    +0.25
    1.2%
    481
    +0.20
    1.0%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    480
    +0.28
    -0.01
    105
    +0.25
    0.03
    119
    +0.20
    0.03
    Negative Logits
    Applications
    -0.77
    izons
    -0.76
    ilitation
    -0.75
    ructure
    -0.73
    ribut
    -0.73
     qualitative
    -0.73
     attainment
    -0.73
    sufficient
    -0.73
    requisite
    -0.72
    egu
    -0.71
    POSITIVE LOGITS
     hilar
    1.16
     hilarious
    1.15
     poop
    1.13
     prank
    1.13
     goof
    1.12
     dudes
    1.11
     dude
    1.11
     jokes
    1.10
     boobs
    1.10
     goofy
    1.08
    Activations Density 1.034%

    No Known Activations