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. 2330
    Prev
    Next
    INDEX
    Explanations

    texts related to open access or open-source initiatives

    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
    7
    +0.11
    0.5%
    535
    +0.09
    0.4%
    240
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    201
    +0.11
    0.04
    240
    +0.09
    0.04
    710
    +0.09
    0.04
    Negative Logits
    iage
    -0.75
    ibur
    -0.65
    ersive
    -0.63
    older
    -0.63
    doms
    -0.63
    CTV
    -0.62
    =-=-=-=-=-=-=-=-
    -0.60
    utical
    -0.60
     Brach
    -0.60
     Commando
    -0.60
    POSITIVE LOGITS
     minded
    0.88
    ended
    0.83
    street
    0.83
    hearted
    0.80
     source
    0.78
    ource
    0.77
    office
    0.77
    SSL
    0.73
     sourced
    0.73
    erness
    0.72
    Activations Density 0.331%

    No Known Activations