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. 43447
    Prev
    Next
    INDEX
    Explanations

    terms related to environmental issues and conservation

    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
    721
    +0.12
    0.5%
    284
    +0.10
    0.5%
    509
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    284
    +0.12
    0.01
    359
    +0.10
    0.01
    354
    +0.09
    0.01
    Negative Logits
    udeb
    -0.75
     antioxid
    -0.70
     CLSID
    -0.68
    ription
    -0.67
     Skydragon
    -0.63
    ellectual
    -0.60
    ettlement
    -0.58
    machine
    -0.58
    road
    -0.57
     fixing
    -0.57
    POSITIVE LOGITS
    theless
    0.97
    rences
    0.92
     Nights
    0.83
     Noct
    0.76
    winter
    0.73
    ited
    0.73
    nces
    0.72
    nce
    0.68
     Never
    0.68
    ounters
    0.67
    Activations Density 0.029%

    No Known Activations