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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsBlogSlackPrivacy & 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. 1230
    Prev
    Next
    INDEX
    Explanations

    words related to investigations or inquiries

    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
    113
    +0.13
    0.6%
    549
    +0.11
    0.5%
    571
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.13
    0.12
    744
    +0.11
    0.09
    540
    +0.11
    0.09
    Negative Logits
    emonic
    -0.74
    birth
    -0.70
     goodbye
    -0.67
    ãĤª
    -0.67
    ane
    -0.62
    icio
    -0.62
    Hero
    -0.61
    Stock
    -0.61
    jc
    -0.60
     prest
    -0.58
    POSITIVE LOGITS
     probing
    1.28
     investigation
    1.12
     investigating
    1.10
     probe
    1.10
     inquiry
    1.05
     investigations
    1.04
     Investigation
    1.04
     hearings
    1.04
     Probe
    1.00
     investigates
    0.99
    Activations Density 1.978%

    No Known Activations