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. 40120
    Prev
    Next
    INDEX
    Explanations

    specific keywords related to societal or political issues

    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
    447
    +0.14
    0.6%
    615
    +0.10
    0.5%
    412
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    224
    +0.14
    0.03
    401
    +0.10
    0.03
    192
    +0.10
    0.02
    Negative Logits
    edIn
    -0.70
    };
    -0.69
    bidden
    -0.69
    externalActionCode
    -0.68
    ca
    -0.66
     Cosponsors
    -0.66
    chest
    -0.65
    Wil
    -0.64
    Cooldown
    -0.64
    iHUD
    -0.64
    POSITIVE LOGITS
     survive
    0.96
     prove
    0.88
     become
    0.86
     succeed
    0.86
     differ
    0.86
     resonate
    0.84
     evolve
    0.83
     weigh
    0.82
     decide
    0.81
     grow
    0.81
    Activations Density 0.132%

    No Known Activations