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 49152
    5. 8-RES_FS49152-JB
    6. 23821
    Prev
    Next
    INDEX
    Explanations

    phrases related to systemic issues and challenges in society

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    +0.51
    2.6%
    266
    +0.13
    0.7%
    447
    +0.13
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    480
    +0.51
    0.01
    266
    +0.13
    0.06
    362
    +0.13
    0.05
    Negative Logits
    nesday
    -0.69
    swer
    -0.62
     accordingly
    -0.61
    Favorite
    -0.61
    wife
    -0.60
    husband
    -0.60
    arettes
    -0.59
    ichever
    -0.59
    submit
    -0.59
    assies
    -0.59
    POSITIVE LOGITS
     systemic
    1.10
     disparities
    1.10
     inequalities
    1.05
     inequ
    1.05
     vulnerabilities
    1.01
     inequality
    1.01
     misconceptions
    0.99
     deficiencies
    0.97
     injust
    0.95
     shortcomings
    0.94
    Activations Density 0.379%

    No Known Activations