© 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 3072
    9. 8-RES_FS3072-JB
    11. 3007
    Prev
    Next
    INDEX
    Explanations

    locations or settings described in a geopolitical context

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    640
    +0.36
    1.8%
    480
    +0.21
    1.0%
    314
    +0.14
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.36
    0.26
    94
    +0.21
    0.17
    143
    +0.14
    0.13
    Negative Logits
     Cent
    -0.67
     dissu
    -0.67
    hement
    -0.66
     diving
    -0.66
    iton
    -0.63
    xual
    -0.63
     wanting
    -0.61
     endeav
    -0.60
     inactive
    -0.60
     representation
    -0.59
    POSITIVE LOGITS
    ³³³³³³³³
    1.19
    ³³³
    1.14
    ³³³³
    1.11
    ³³³³³³³³³³³³³³³³
    1.07
    Posted
    1.01
    ³³
    0.89
    Contents
    0.87
    ramid
    0.85
    posted
    0.84
    CVE
    0.84
    Activations Density 1.089%

    No Known Activations