© 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 24576
    9. 8-RES_FS24576-JB
    11. 8695
    Prev
    Next
    INDEX
    Explanations

    political and social terms related to international affairs and policies

    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
    288
    +0.16
    0.8%
    18
    +0.14
    0.6%
    362
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.16
    0.05
    18
    +0.14
    0.05
    404
    +0.11
    0.04
    Negative Logits
    ĸļ
    -1.14
    ciating
    -0.85
    llular
    -0.82
    emis
    -0.82
    ricanes
    -0.82
    cedented
    -0.82
     Drawn
    -0.81
    aunder
    -0.81
    avers
    -0.81
    dropping
    -0.81
    POSITIVE LOGITS
     tendencies
    1.29
     zeal
    1.18
     ferv
    1.13
     ideals
    1.11
     sentiments
    1.10
     ideology
    1.09
     impulses
    1.08
     worldview
    1.05
     sensibilities
    1.05
     ethos
    1.03
    Activations Density 0.117%

    No Known Activations