© 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. 19885
    Prev
    Next
    INDEX
    Explanations

    time-related phrases like "over the next few years"

    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
    175
    +0.10
    0.5%
    641
    +0.10
    0.4%
    269
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    175
    +0.10
    0.03
    223
    +0.10
    0.03
    199
    +0.09
    0.03
    Negative Logits
    osure
    -0.69
    apt
    -0.68
    awan
    -0.67
    iquid
    -0.63
     Franch
    -0.63
    bley
    -0.63
     Ahead
    -0.62
    VILLE
    -0.62
    iard
    -0.62
    ioned
    -0.61
    POSITIVE LOGITS
     decade
    1.28
     decades
    1.14
    ebin
    1.05
     millennia
    1.03
     weeks
    1.03
     months
    1.01
     years
    1.00
     generations
    1.00
     fortnight
    1.00
     centuries
    0.98
    Activations Density 0.047%

    No Known Activations