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. 43180
    Prev
    Next
    INDEX
    Explanations

    occurrences of specific names and titles in a context involving arrest or law enforcement

    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
    481
    +0.17
    0.8%
    597
    +0.12
    0.6%
    482
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.17
    0.03
    659
    +0.12
    0.03
    392
    +0.11
    0.02
    Negative Logits
     Tyr
    -0.89
     Rece
    -0.81
     Depot
    -0.77
     Tomato
    -0.75
     148
    -0.74
     tyr
    -0.73
     Betty
    -0.73
    $$$$
    -0.68
     sher
    -0.68
     predicate
    -0.67
    POSITIVE LOGITS
    ong
    1.51
    ang
    1.42
    ongs
    1.23
    ANG
    1.15
    angs
    1.09
    ONG
    1.07
    onga
    1.07
    yang
    1.04
    ung
    1.00
    ANGE
    0.99
    Activations Density 0.069%

    No Known Activations