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. 43218
    Prev
    Next
    INDEX
    Explanations

    information related to official registrations and statistics

    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
    +0.48
    2.4%
    266
    +0.13
    0.7%
    87
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    480
    +0.48
    0.02
    266
    +0.13
    0.05
    326
    +0.13
    0.05
    Negative Logits
     corrid
    -0.68
     psychiat
    -0.62
     Adin
    -0.60
    )."
    -0.59
     undermin
    -0.59
     conflic
    -0.57
     surrog
    -0.57
    \)
    -0.57
     neighb
    -0.56
     explan
    -0.55
    POSITIVE LOGITS
     âĵĺ
    0.87
     IMAGES
    0.75
     Female
    0.75
     Male
    0.74
    Joined
    0.71
     Discuss
    0.65
    Female
    0.62
    Topic
    0.62
     Gender
    0.61
    Male
    0.60
    Activations Density 0.335%

    No Known Activations