© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 24576
    9. 8-RES_FS24576-JB
    11. 22099
    Prev
    Next
    INDEX
    Explanations

    terms related to scientific measurements and their implications, particularly in experimental contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    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
    575
    +0.11
    0.5%
    358
    +0.11
    0.5%
    336
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    575
    +0.11
    0.11
    288
    +0.11
    0.11
    762
    +0.10
    0.08
    Negative Logits
     Polit
    -0.69
     Samar
    -0.65
    Weather
    -0.63
     Jur
    -0.61
    ãĥ£
    -0.57
    ortium
    -0.57
     Fiscal
    -0.56
    xtap
    -0.53
    cohol
    -0.53
     Investigation
    -0.53
    POSITIVE LOGITS
    +.
    0.88
     altogether
    0.83
     anyways
    0.79
     anyway
    0.78
    .'
    0.77
    .</
    0.76
    .'"
    0.76
    .�
    0.76
     herself
    0.75
     anymore
    0.75
    Activations Density 0.780%

    No Known Activations