© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 41531
    Prev
    Next
    INDEX
    Explanations

    specific numerical values, particularly related to years and statistics

    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
    396
    +0.10
    0.5%
    189
    +0.10
    0.5%
    462
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    310
    +0.10
    0.04
    220
    +0.10
    0.04
    260
    +0.10
    0.04
    Negative Logits
    LOAD
    -0.79
    ei
    -0.78
    hner
    -0.78
    gie
    -0.76
    eer
    -0.74
    icz
    -0.72
    geist
    -0.71
    mes
    -0.69
    iard
    -0.69
    izen
    -0.69
    POSITIVE LOGITS
     ILCS
    1.32
    %:
    0.86
    ERSON
    0.82
    %-
    0.77
     Yard
    0.76
    ISH
    0.75
    %
    0.73
    ishing
    0.73
    00
    0.73
    ances
    0.72
    Activations Density 0.077%

    No Known Activations