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. 341
    Prev
    Next
    INDEX
    Explanations

    phrases related to social issues and challenges, such as poverty, homelessness, discrimination, and lack of access to resources

    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
    480
    +0.42
    2.1%
    86
    +0.09
    0.5%
    420
    +0.09
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    480
    +0.42
    0.00
    724
    +0.09
    0.07
    150
    +0.09
    0.04
    Negative Logits
    £ı
    -0.76
    ocular
    -0.74
    afort
    -0.69
     Cannes
    -0.69
    canon
    -0.69
    iple
    -0.67
    eus
    -0.65
     nods
    -0.65
    poke
    -0.65
     handshake
    -0.65
    POSITIVE LOGITS
     poverty
    1.38
     unemployment
    1.31
     malnutrition
    1.29
     impover
    1.28
     homelessness
    1.23
     Unemployment
    1.20
     worsening
    1.16
     starvation
    1.15
     overcrowd
    1.14
     shortages
    1.13
    Activations Density 0.853%

    No Known Activations