© 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. 26516
    Prev
    Next
    INDEX
    Explanations

    repeated requests or mentions of a website or visit

    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
    182
    +0.12
    0.5%
    373
    +0.11
    0.5%
    604
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    182
    +0.12
    0.01
    604
    +0.11
    0.01
    643
    +0.10
    0.01
    Negative Logits
    paces
    -0.78
     receptive
    -0.68
     sten
    -0.66
    cil
    -0.66
    pire
    -0.64
     cub
    -0.64
     polymorph
    -0.63
    NetMessage
    -0.60
    plates
    -0.60
     subp
    -0.60
    POSITIVE LOGITS
    acion
    1.01
    osity
    0.74
    yon
    0.69
     Website
    0.69
     Visit
    0.68
    ogly
    0.66
    aghan
    0.65
     Hover
    0.62
    æĸ¹
    0.62
     Subscribe
    0.61
    Activations Density 0.007%

    No Known Activations