Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 49152
    5. 8-RES_FS49152-JB
    6. 40085
    Prev
    Next
    INDEX
    Explanations

    occurrences of the letter combination "el" in various forms

    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
    +0.16
    0.7%
    373
    +0.15
    0.7%
    721
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    373
    +0.16
    -0.00
    522
    +0.15
    0.01
    284
    +0.10
    0.01
    Negative Logits
     redundancy
    -0.67
     favour
    -0.65
     toler
    -0.65
     constraint
    -0.64
     constraints
    -0.60
     sideline
    -0.59
     stagnation
    -0.59
     designing
    -0.59
     portfolios
    -0.57
     suspensions
    -0.57
    POSITIVE LOGITS
    phia
    1.12
    phi
    1.02
    aida
    0.96
    bert
    0.95
    anto
    0.92
    rican
    0.90
    ph
    0.85
    zin
    0.83
    icia
    0.81
    ici
    0.80
    Activations Density 0.010%

    No Known Activations