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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & 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. 43370
    Prev
    Next
    INDEX
    Explanations

    terms associated with influence and definition

    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
    403
    +0.13
    0.6%
    408
    +0.12
    0.5%
    691
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    403
    +0.13
    0.03
    22
    +0.12
    0.03
    123
    +0.11
    0.03
    Negative Logits
    iphate
    -0.71
    DERR
    -0.70
    idel
    -0.69
    ighters
    -0.69
    igers
    -0.67
    ãĥı
    -0.67
    intend
    -0.67
    aeus
    -0.65
    sels
    -0.64
    leased
    -0.64
    POSITIVE LOGITS
     contemporary
    0.85
     everything
    0.80
     our
    0.78
     discussions
    0.78
     debates
    0.77
     both
    0.75
     modern
    0.74
     anthropology
    0.72
     colonialism
    0.72
     many
    0.71
    Activations Density 0.071%

    No Known Activations