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. 37119
    Prev
    Next
    INDEX
    Explanations

    instances of the word "perspective" and its variations, indicating it is looking for discussions around viewpoints or angles on various topics

    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
    195
    +0.17
    0.8%
    575
    +0.12
    0.6%
    641
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    195
    +0.17
    0.01
    575
    +0.12
    0.01
    641
    +0.10
    0.01
    Negative Logits
    Interstitial
    -0.83
    liam
    -0.76
    nard
    -0.76
    ertodd
    -0.73
    avis
    -0.72
    ãĥīãĥ©
    -0.70
    atron
    -0.70
    duct
    -0.68
    vez
    -0.68
    vous
    -0.67
    POSITIVE LOGITS
     perspectives
    1.01
     perspective
    0.98
     viewpoint
    0.90
     viewpoints
    0.88
    spection
    0.86
     views
    0.83
     view
    0.80
     Perspective
    0.80
     lens
    0.78
     Lens
    0.69
    Activations Density 0.013%

    No Known Activations