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

    Neuronpedia

    Jacobian LensNEW
    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. 43115
    Prev
    Next
    INDEX
    Explanations

    numbers, particularly those related to counting or quantifying

    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
    314
    +0.11
    0.5%
    344
    +0.11
    0.5%
    71
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    450
    +0.11
    0.02
    223
    +0.11
    0.02
    260
    +0.11
    0.02
    Negative Logits
    Activity
    -0.78
    Bank
    -0.77
    Helper
    -0.62
     Redemption
    -0.60
    Climate
    -0.60
     Caption
    -0.60
    aque
    -0.60
    ller
    -0.59
    ghai
    -0.59
    Guard
    -0.58
    POSITIVE LOGITS
     thousand
    1.26
    eenth
    1.07
     Thousand
    0.93
    something
    0.92
    ousand
    0.92
    fold
    0.90
     hundred
    0.89
    ieth
    0.84
    ements
    0.80
     eighty
    0.80
    Activations Density 0.016%

    No Known Activations