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 1536
    5. 8-RES_FS1536-JB
    6. 1333
    Prev
    Next
    INDEX
    Explanations

    real estate-related terms and phrases

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    642
    +0.14
    0.6%
    152
    +0.12
    0.6%
    167
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.14
    0.16
    152
    +0.12
    0.11
    575
    +0.10
    0.11
    Negative Logits
    xon
    -0.88
    Downloadha
    -0.81
    otine
    -0.77
    nir
    -0.72
    £ı
    -0.69
     paralysis
    -0.68
    aneously
    -0.68
    gans
    -0.68
    zyk
    -0.67
    auts
    -0.66
    POSITIVE LOGITS
     apartments
    1.21
     tenants
    1.21
     dwellings
    1.11
     apartment
    1.10
     tenant
    1.10
     renters
    1.09
     estate
    1.09
     housing
    1.08
     occupancy
    1.05
     estates
    1.05
    Activations Density 2.440%

    No Known Activations