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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 12288
    9. 8-RES_FS12288-JB
    11. 10151
    Prev
    Next
    INDEX
    Explanations

    words related to medical or health crises or emergencies

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    373
    +0.13
    0.6%
    444
    +0.10
    0.4%
    109
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    89
    +0.13
    0.01
    444
    +0.10
    0.01
    455
    +0.09
    0.02
    Negative Logits
    wit
    -0.77
    commit
    -0.70
    iewicz
    -0.69
    lite
    -0.68
    stood
    -0.67
    bush
    -0.67
     fame
    -0.65
    storms
    -0.64
    bats
    -0.61
    naire
    -0.61
    POSITIVE LOGITS
     unsupported
    1.02
    alysed
    0.72
    Ĭ
    0.69
     BELOW
    0.68
     prohibited
    0.67
     supplied
    0.67
    optional
    0.67
    Streamer
    0.67
     playback
    0.63
     not
    0.63
    Activations Density 0.059%

    No Known Activations