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 24576
    5. 8-RES_FS24576-JB
    6. 23643
    Prev
    Next
    INDEX
    Explanations

    terms related to gliding and various aspects of gliding activities and communities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    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
    447
    +0.14
    0.6%
    575
    +0.10
    0.5%
    153
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    575
    +0.14
    0.05
    537
    +0.10
    0.04
    288
    +0.09
    0.05
    Negative Logits
     restruct
    -0.81
     Baghd
    -0.73
     reinvest
    -0.72
     Diss
    -0.69
     substantive
    -0.69
    ãĥ¤
    -0.68
    otten
    -0.68
     redacted
    -0.65
    metics
    -0.65
    Process
    -0.64
    POSITIVE LOGITS
     biking
    1.15
     instructor
    1.09
     bikes
    1.08
     bicycles
    1.06
     canoe
    1.05
     racer
    1.04
     skiing
    1.04
     padd
    1.02
     motorcycles
    1.00
     sled
    0.99
    Activations Density 0.196%

    No Known Activations