© 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 1536
    9. 8-RES_FS1536-JB
    11. 68
    Prev
    Next
    INDEX
    Explanations

    keywords related to political issues and debates

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    737
    +0.14
    0.6%
    645
    +0.10
    0.5%
    714
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    714
    +0.14
    0.07
    737
    +0.10
    0.08
    507
    +0.10
    0.09
    Negative Logits
    Synopsis
    -0.75
    ascript
    -0.73
    arta
    -0.73
    Closure
    -0.70
    BILL
    -0.68
    acas
    -0.66
    REP
    -0.63
    soType
    -0.63
    binding
    -0.63
    chemy
    -0.62
    POSITIVE LOGITS
     weakest
    1.02
     lowest
    1.00
     narrower
    0.99
     favoured
    0.99
     predomin
    0.98
     weaker
    0.98
     favored
    0.96
     intermediate
    0.94
     preferred
    0.94
     outl
    0.93
    Activations Density 4.767%

    No Known Activations