© 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 49152
    9. 8-RES_FS49152-JB
    11. 41720
    Prev
    Next
    INDEX
    Explanations

    words related to dominance or control in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    447
    +0.16
    0.7%
    271
    +0.11
    0.5%
    288
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    271
    +0.16
    0.01
    517
    +0.11
    0.01
    143
    +0.10
    0.01
    Negative Logits
    resso
    -0.72
    leeve
    -0.69
    endment
    -0.67
    phans
    -0.66
    atana
    -0.65
    spir
    -0.64
    osta
    -0.64
    ullah
    -0.64
    ricanes
    -0.63
    DragonMagazine
    -0.63
    POSITIVE LOGITS
     headlines
    1.05
     dominated
    0.69
     proceedings
    0.67
     overshadowed
    0.65
     dominates
    0.65
     charge
    0.65
    pread
    0.64
     shaping
    0.62
    quez
    0.61
    ively
    0.61
    Activations Density 0.017%

    No Known Activations