© 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 24576
    9. 8-RES_FS24576-JB
    11. 9431
    Prev
    Next
    INDEX
    Explanations

    articles or discussions related to legal matters or legislation

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    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
    718
    +0.14
    0.7%
    755
    +0.10
    0.5%
    608
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    718
    +0.14
    0.05
    49
    +0.10
    0.04
    755
    +0.10
    0.03
    Negative Logits
     nets
    -0.62
    dro
    -0.59
     Appears
    -0.58
    lot
    -0.57
     dotted
    -0.57
    pict
    -0.57
     pic
    -0.56
    calling
    -0.56
     forg
    -0.56
    hops
    -0.55
    POSITIVE LOGITS
    ilet
    1.41
     summarize
    1.25
     reiterate
    1.16
     illustrate
    1.13
     complicate
    1.10
    ppings
    1.07
     clarify
    1.07
     compensate
    1.06
     paraph
    1.04
    pping
    1.03
    Activations Density 0.055%

    No Known Activations