© 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 24576
    9. 8-RES_FS24576-JB
    11. 9396
    Prev
    Next
    INDEX
    Explanations

    phrases related to statistics and data analysis

    oai_token-act-pair · gpt-3.5-turboTriggered 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
    216
    +0.14
    0.7%
    653
    +0.11
    0.5%
    31
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    216
    +0.14
    0.05
    653
    +0.11
    0.04
    623
    +0.10
    0.03
    Negative Logits
    emis
    -0.74
    gnu
    -0.72
    çīĪ
    -0.68
    rology
    -0.66
     Nich
    -0.63
    vice
    -0.62
    onto
    -0.62
    onso
    -0.62
    ogens
    -0.62
    ania
    -0.61
    POSITIVE LOGITS
     quantities
    1.13
     scale
    1.11
     sized
    1.10
     intestine
    1.09
     (>
    1.06
     amounts
    1.05
     swat
    0.96
     enough
    0.92
     quantity
    0.91
     sums
    0.91
    Activations Density 0.064%

    No Known Activations