© 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. 37315
    Prev
    Next
    INDEX
    Explanations

    terms associated with global or widespread events and issues

    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
    554
    +0.12
    0.5%
    591
    +0.10
    0.4%
    414
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    554
    +0.12
    0.02
    288
    +0.10
    0.04
    445
    +0.09
    0.03
    Negative Logits
    mate
    -0.67
     Mn
    -0.67
     contained
    -0.64
    ertodd
    -0.62
    astery
    -0.62
    axis
    -0.60
    tein
    -0.60
     Constitution
    -0.60
     manually
    -0.59
     expires
    -0.59
    POSITIVE LOGITS
     acclaim
    1.39
     attention
    1.29
     outcry
    1.25
     ridicule
    1.15
     scorn
    1.15
     condemnation
    1.10
     uproar
    1.08
     notoriety
    1.06
     outrage
    1.05
     criticism
    1.02
    Activations Density 0.103%

    No Known Activations