© 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 12288
    9. 8-RES_FS12288-JB
    11. 10525
    Prev
    Next
    INDEX
    Explanations

    statistical data, legal terms, and policy-related keywords

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    447
    +0.12
    0.6%
    430
    +0.12
    0.6%
    575
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    575
    +0.12
    0.10
    51
    +0.12
    0.09
    660
    +0.11
    0.06
    Negative Logits
    mosp
    -0.78
    rawdownloadcloneembedreportprint
    -0.78
    Oracle
    -0.74
     Kernel
    -0.67
    ãĥķãĤ©
    -0.67
    SourceFile
    -0.67
    Revolution
    -0.66
    OLOG
    -0.65
    UTERS
    -0.64
     Samurai
    -0.64
    POSITIVE LOGITS
     who
    1.18
     enrolled
    1.11
     accessing
    1.10
     aged
    1.08
     undergoing
    1.04
     residing
    1.04
     fleeing
    1.02
     eligible
    0.99
     experiencing
    0.99
     transitioning
    0.98
    Activations Density 0.507%

    No Known Activations