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. 924
    Prev
    Next
    INDEX
    Explanations

    phrases related to questioning societal norms or beliefs

    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
    412
    +0.10
    0.5%
    615
    +0.10
    0.4%
    414
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    687
    +0.10
    0.06
    638
    +0.10
    0.06
    449
    +0.10
    0.05
    Negative Logits
    inery
    -0.74
    iHUD
    -0.73
    ©¶æ¥µ
    -0.71
    igi
    -0.71
    APD
    -0.70
    abor
    -0.69
    DERR
    -0.69
    ullivan
    -0.68
    catentry
    -0.67
    abba
    -0.65
    POSITIVE LOGITS
    ?
    2.18
    ?:
    2.06
    ?"
    1.97
    ?)
    1.93
    ?'
    1.93
    ?).
    1.91
    ?),
    1.91
    ?",
    1.89
    ?!
    1.85
    ?,
    1.82
    Activations Density 1.405%

    No Known Activations