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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 1536
    5. 8-RES_FS1536-JB
    6. 550
    Prev
    Next
    INDEX
    Explanations

    terms and phrases related to software versions and updates

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    393
    +0.13
    0.6%
    332
    +0.10
    0.4%
    549
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.13
    0.17
    608
    +0.10
    0.17
    393
    +0.09
    0.15
    Negative Logits
    urers
    -0.72
    arers
    -0.72
    lees
    -0.72
    entimes
    -0.69
    ctuary
    -0.69
    ombies
    -0.66
    rals
    -0.65
    cks
    -0.64
    eneg
    -0.64
    rings
    -0.63
    POSITIVE LOGITS
     sorts
    0.75
     the
    0.75
     theirs
    0.72
     SCP
    0.69
     Camel
    0.66
     hers
    0.63
    ©¶æ
    0.61
     Wonderland
    0.60
     ours
    0.60
     oneself
    0.57
    Activations Density 2.990%

    No Known Activations