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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 25843
    Prev
    Next
    INDEX
    Explanations

    instances of the string "igan."

    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
    659
    +0.12
    0.6%
    587
    +0.11
    0.5%
    235
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.12
    0.01
    235
    +0.11
    0.01
    587
    +0.11
    0.01
    Negative Logits
    dates
    -0.74
    nesday
    -0.73
     Seym
    -0.66
    ngth
    -0.65
    ensions
    -0.64
    ĨĴ
    -0.64
    lems
    -0.63
    ¥µ
    -0.63
    grades
    -0.63
    itudes
    -0.62
    POSITIVE LOGITS
    igans
    1.58
    igan
    1.34
    omics
    0.93
    osa
    0.85
    owski
    0.80
    ieri
    0.80
    otos
    0.76
    ator
    0.76
    cheon
    0.75
    batch
    0.75
    Activations Density 0.005%

    No Known Activations