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 3072
    5. 8-RES_FS3072-JB
    6. 2955
    Prev
    Next
    INDEX
    Explanations

    phrases related to real-world situations or scenarios

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    642
    +0.13
    0.6%
    152
    +0.11
    0.5%
    314
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    642
    +0.13
    0.05
    61
    +0.11
    0.04
    493
    +0.09
    0.04
    Negative Logits
    xual
    -0.95
    bard
    -0.65
    osi
    -0.64
     Include
    -0.62
     Vaugh
    -0.61
     Carbuncle
    -0.60
    edin
    -0.60
    rav
    -0.60
    azard
    -0.60
    ansk
    -0.59
    POSITIVE LOGITS
    ignment
    1.45
     estate
    1.35
    isation
    1.31
    polit
    1.24
    estate
    1.18
    igned
    1.16
    izations
    1.15
    igning
    1.15
    istically
    1.13
    izable
    1.13
    Activations Density 0.341%

    No Known Activations