© 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 49152
    9. 8-RES_FS49152-JB
    11. 41795
    Prev
    Next
    INDEX
    Explanations

    references to charters and notable individuals associated with them

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    288
    +0.18
    0.9%
    359
    +0.12
    0.6%
    178
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.18
    0.01
    359
    +0.12
    0.01
    343
    +0.11
    0.01
    Negative Logits
    Downloadha
    -0.75
    age
    -0.70
    ential
    -0.68
    EMS
    -0.67
    pez
    -0.66
    encer
    -0.66
    nesota
    -0.65
    afort
    -0.65
    uana
    -0.64
    eah
    -0.64
    POSITIVE LOGITS
    onic
    0.85
     Framework
    0.80
    izon
    0.75
    onite
    0.74
    zman
    0.74
     Amendment
    0.73
    jee
    0.71
    witz
    0.71
    istic
    0.68
     Amendments
    0.68
    Activations Density 0.004%

    No Known Activations