© 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. 7303
    Prev
    Next
    INDEX
    Explanations

    phrases related to religious beliefs and community initiatives

    oai_token-act-pair · gpt-3.5-turboTriggered 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
    101
    +0.12
    0.5%
    6
    +0.10
    0.5%
    160
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    6
    +0.12
    0.03
    48
    +0.10
    0.02
    166
    +0.10
    0.02
    Negative Logits
    iage
    -0.79
     satell
    -0.74
    ijah
    -0.73
    izzard
    -0.73
     mathemat
    -0.73
     glim
    -0.71
     loft
    -0.69
     fortun
    -0.67
    bris
    -0.65
     Companion
    -0.62
    POSITIVE LOGITS
    âĶĢâĶĢâĶĢâĶĢ
    1.10
    """
    1.08
    âĢ
    1.06
    âĶĢâĶĢ
    0.99
    %"
    0.96
    ¯
    0.94
    ï¸ı
    0.93
    >>>
    0.89
    \/
    0.86
    »
    0.85
    Activations Density 0.050%

    No Known Activations