© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 1866
    Prev
    Next
    INDEX
    Explanations

    phrases related to reasons or causation

    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
    679
    +0.13
    0.6%
    218
    +0.12
    0.6%
    738
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    65
    +0.13
    0.04
    218
    +0.12
    0.03
    738
    +0.10
    0.03
    Negative Logits
    lem
    -0.78
    intern
    -0.69
    ymph
    -0.69
    Gas
    -0.68
    age
    -0.67
    agin
    -0.65
    mint
    -0.65
    åĤ
    -0.64
    ses
    -0.64
     exting
    -0.64
    POSITIVE LOGITS
    rely
    0.88
     there
    0.79
     nobody
    0.79
     they
    0.73
     unlike
    0.72
     we
    0.70
    ecause
    0.68
     it
    0.67
     of
    0.65
     hindsight
    0.64
    Activations Density 0.096%

    No Known Activations