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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 23884
    Prev
    Next
    INDEX
    Explanations

    words related to escalating actions or conditions, particularly in a negative context

    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
    230
    +0.12
    0.5%
    343
    +0.09
    0.4%
    187
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    230
    +0.12
    0.02
    240
    +0.09
    0.01
    251
    +0.09
    0.01
    Negative Logits
    DragonMagazine
    -0.88
    soDeliveryDate
    -0.73
    wic
    -0.70
    izon
    -0.69
    iverpool
    -0.67
     Territories
    -0.64
     Leilan
    -0.62
    tical
    -0.62
    é¾įå
    -0.60
    roma
    -0.59
    POSITIVE LOGITS
     up
    1.12
    Up
    0.96
     Up
    0.93
    up
    0.91
     UP
    0.82
    ups
    0.80
     upward
    0.80
     leaps
    0.80
     ups
    0.76
     exponentially
    0.72
    Activations Density 0.027%

    No Known Activations