Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 25823
    Prev
    Next
    INDEX
    Explanations

    terms related to legal, educational, and market contexts

    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
    481
    +0.14
    0.6%
    625
    +0.13
    0.6%
    288
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.14
    0.07
    575
    +0.13
    0.06
    325
    +0.12
    0.05
    Negative Logits
    earchers
    -0.70
    ottest
    -0.67
    aughtered
    -0.65
     Sections
    -0.65
     Klux
    -0.64
     largeDownload
    -0.63
     copies
    -0.62
    ositories
    -0.61
    osponsors
    -0.61
     advis
    -0.60
    POSITIVE LOGITS
    pour
    0.79
     involving
    0.79
    sword
    0.75
    ifier
    0.74
     akin
    0.73
     like
    0.73
     devoid
    0.72
     called
    0.72
    .#
    0.72
     whose
    0.72
    Activations Density 0.277%

    No Known Activations