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. 1183
    Prev
    Next
    INDEX
    Explanations

    instances of the phrase "too often" in the text

    oai_token-act-pair · gpt-3.5-turboTriggered 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
    481
    +0.22
    1.0%
    447
    +0.14
    0.6%
    58
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    481
    +0.22
    0.07
    34
    +0.14
    0.03
    170
    +0.11
    0.04
    Negative Logits
    bol
    -0.72
     Sever
    -0.64
    uddin
    -0.64
    orah
    -0.62
     TNT
    -0.61
    ebus
    -0.60
     Panzer
    -0.60
    stroke
    -0.59
    elta
    -0.59
    orthy
    -0.59
    POSITIVE LOGITS
     misconceptions
    1.05
     outdated
    1.00
     stereotypes
    0.97
     societal
    0.96
     stigma
    0.91
     industrialized
    0.90
     stereotype
    0.88
     ignorance
    0.88
     society
    0.88
     misinformation
    0.87
    Activations Density 0.817%

    No Known Activations