© 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 24576
    9. 8-RES_FS24576-JB
    11. 23894
    Prev
    Next
    INDEX
    Explanations

    phrases indicating criticism or concern regarding social and political issues

    oai_token-act-pair · gpt-4o-miniTriggered 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
    373
    +0.12
    0.6%
    314
    +0.11
    0.5%
    333
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    333
    +0.12
    0.07
    675
    +0.11
    0.07
    168
    +0.11
    0.07
    Negative Logits
     lod
    -0.76
     trail
    -0.72
     reflex
    -0.70
     posit
    -0.70
     lapse
    -0.69
     metic
    -0.67
     chained
    -0.67
     territorial
    -0.67
     pigeon
    -0.66
     rack
    -0.66
    POSITIVE LOGITS
    ï¸ı
    1.18
    ternity
    1.02
    there
    0.94
    prep
    0.90
    audi
    0.86
    ×Ķ
    0.86
    www
    0.83
    uthor
    0.81
    ever
    0.81
    \-
    0.81
    Activations Density 0.142%

    No Known Activations