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. 12153
    Prev
    Next
    INDEX
    Explanations

    terms related to declarations of emergencies

    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
    288
    +0.13
    0.6%
    86
    +0.11
    0.5%
    572
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    132
    +0.13
    0.01
    606
    +0.11
    0.01
    572
    +0.10
    0.01
    Negative Logits
    ewski
    -0.83
    Ģ
    -0.83
    kies
    -0.81
    eday
    -0.79
    Ķ
    -0.78
    sburgh
    -0.78
    ģ
    -0.77
    guyen
    -0.77
    hetti
    -0.77
    models
    -0.73
    POSITIVE LOGITS
     decree
    0.97
     truce
    0.91
     liquidity
    0.90
     proclamation
    0.88
     declaration
    0.83
     ceasefire
    0.82
     drought
    0.81
     measures
    0.80
     moratorium
    0.80
     brake
    0.80
    Activations Density 0.009%

    No Known Activations