© 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 12288
    9. 8-RES_FS12288-JB
    11. 10321
    Prev
    Next
    INDEX
    Explanations

    references to notable historical figures or events

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    98
    +0.11
    0.5%
    373
    +0.11
    0.5%
    0
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.11
    0.05
    373
    +0.11
    -0.02
    701
    +0.10
    0.04
    Negative Logits
    ãĤ´ãĥ³
    -0.76
    ĵĺ
    -0.73
     ILCS
    -0.72
     skinny
    -0.63
     corrosion
    -0.62
     responsiveness
    -0.61
     fries
    -0.60
     neutrality
    -0.59
     camoufl
    -0.57
     Arrows
    -0.57
    POSITIVE LOGITS
    arb
    0.81
    pora
    0.80
    andowski
    0.79
    weekly
    0.78
    incial
    0.72
    idis
    0.72
    etus
    0.71
    cious
    0.70
    ache
    0.70
    ulo
    0.69
    Activations Density 0.289%

    No Known Activations