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. 1895
    Prev
    Next
    INDEX
    Explanations

    text related to sharing information or articles

    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
    314
    +0.19
    0.9%
    480
    +0.15
    0.7%
    176
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    267
    +0.19
    0.10
    119
    +0.15
    0.09
    273
    +0.10
    0.10
    Negative Logits
    .","
    -0.71
     liqu
    -0.66
    itiz
    -0.65
    -"
    -0.65
     thereto
    -0.64
    >>>>>>>>
    -0.63
     ..."
    -0.63
     ``(
    -0.60
    "},"
    -0.59
     accordingly
    -0.59
    POSITIVE LOGITS
    resa
    1.52
    odore
    1.24
    xiety
    1.10
    anmar
    0.99
    alyst
    0.99
    bidden
    0.98
    mosp
    0.97
    swers
    0.93
    cloneembedreportprint
    0.93
    pherd
    0.91
    Activations Density 1.896%

    No Known Activations