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 1536
    5. 8-RES_FS1536-JB
    6. 800
    Prev
    Next
    INDEX
    Explanations

    phrases related to political and governmental issues

    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
    570
    +0.14
    0.6%
    459
    +0.10
    0.5%
    165
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    570
    +0.14
    0.04
    762
    +0.10
    0.21
    756
    +0.10
    -0.11
    Negative Logits
    ãĤ´ãĥ³
    -0.91
    surprisingly
    -0.65
    translation
    -0.59
    ©¶æ¥µ
    -0.58
     ):
    -0.57
     Whedon
    -0.56
     "#
    -0.56
     "$
    -0.55
    aired
    -0.54
     NFC
    -0.52
    POSITIVE LOGITS
     ..."
    2.10
     â̦"
    1.98
    )."
    1.75
    ),"
    1.74
    .")
    1.74
    )",
    1.65
    )"
    1.65
    ,'"
    1.65
    â̦"
    1.64
    ..."
    1.63
    Activations Density 8.081%

    No Known Activations