© 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 1536
    9. 8-RES_FS1536-JB
    11. 15
    Prev
    Next
    INDEX
    Explanations

    dates and events in the form of month and year

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    361
    +0.11
    0.5%
    139
    +0.11
    0.5%
    633
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    460
    +0.11
    0.08
    361
    +0.11
    0.08
    651
    +0.09
    0.10
    Negative Logits
     Cosponsors
    -0.82
    ãĤ¨ãĥ«
    -0.71
     meshes
    -0.69
     affinity
    -0.68
    anooga
    -0.68
    ventus
    -0.68
    onyms
    -0.67
    oyal
    -0.65
    ãĥĦ
    -0.63
    ractive
    -0.63
    POSITIVE LOGITS
     anniversary
    1.18
    iversary
    1.08
     holidays
    1.03
     reckoning
    1.03
     commem
    0.98
     Anniversary
    0.98
    stice
    0.95
     timeframe
    0.93
     calendar
    0.93
     countdown
    0.91
    Activations Density 4.507%

    No Known Activations