© 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. 508
    Prev
    Next
    INDEX
    Explanations

    phrases related to large quantity or numbers

    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
    383
    +0.11
    0.5%
    241
    +0.11
    0.5%
    71
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    596
    +0.11
    0.03
    241
    +0.11
    0.03
    71
    +0.11
    0.02
    Negative Logits
     Advocate
    -0.84
    SPONSORED
    -0.82
    heid
    -0.76
    oute
    -0.67
    Adv
    -0.66
    arthed
    -0.64
    igent
    -0.62
    âĢ¢âĢ¢
    -0.61
    HCR
    -0.61
     Nich
    -0.61
    POSITIVE LOGITS
    tery
    1.10
    ãĤ§
    0.89
    ortment
    0.80
    creen
    0.77
    rats
    0.76
    ãĥīãĥ©
    0.76
    oys
    0.75
    uay
    0.75
    ums
    0.75
    uses
    0.75
    Activations Density 0.064%

    No Known Activations