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 49152
    5. 8-RES_FS49152-JB
    6. 26289
    Prev
    Next
    INDEX
    Explanations

    concepts related to social consequences and dynamics

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    756
    +0.17
    0.8%
    481
    +0.14
    0.7%
    762
    +0.13
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    762
    +0.17
    0.09
    51
    +0.14
    0.06
    225
    +0.13
    0.07
    Negative Logits
    ©¶æ
    -0.81
    ovember
    -0.69
     )]
    -0.63
    ursday
    -0.63
     Canaver
    -0.61
    DragonMagazine
    -0.60
    çīĪ
    -0.59
    itled
    -0.59
     ?)
    -0.59
    igham
    -0.59
    POSITIVE LOGITS
    .
    0.99
     anyway
    0.89
     wherever
    0.89
     because
    0.87
     regardless
    0.86
    ;
    0.85
    .[
    0.81
     whereas
    0.80
     lest
    0.80
     unless
    0.80
    Activations Density 0.578%

    No Known Activations