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 3072
    5. 8-RES_FS3072-JB
    6. 1993
    Prev
    Next
    INDEX
    Explanations

    phone-related terms

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    289
    +0.12
    0.6%
    213
    +0.11
    0.5%
    352
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    289
    +0.12
    0.03
    352
    +0.11
    0.05
    326
    +0.10
    0.05
    Negative Logits
    rix
    -0.79
    alis
    -0.72
    iverpool
    -0.67
    idth
    -0.63
    itch
    -0.63
    ivia
    -0.63
    utz
    -0.62
    rict
    -0.62
    ivari
    -0.61
     Drac
    -0.61
    POSITIVE LOGITS
     handset
    0.97
    phones
    0.95
     phones
    0.90
     charger
    0.89
     phone
    0.85
    ysis
    0.85
    phone
    0.83
    Numbers
    0.82
     booth
    0.81
     ringing
    0.81
    Activations Density 0.541%

    No Known Activations