© 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
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 12288
    5. 8-RES_FS12288-JB
    6. 9570
    Prev
    Next
    INDEX
    Explanations

    code snippets related to database query and configuration

    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
    50
    +0.12
    0.6%
    210
    +0.11
    0.5%
    696
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.12
    0.03
    289
    +0.11
    0.01
    208
    +0.11
    0.03
    Negative Logits
     scrut
    -0.73
    hower
    -0.66
    enegger
    -0.64
     Yel
    -0.63
    é¾įå
    -0.62
     Aires
    -0.62
    ournal
    -0.61
     Bam
    -0.61
    ecause
    -0.61
    sed
    -0.60
    POSITIVE LOGITS
    _
    0.99
    __
    0.95
    _-_
    0.87
    "}
    0.86
    "},
    0.76
    ">
    0.74
    }}
    0.72
    ___
    0.71
    riors
    0.69
    ._
    0.69
    Activations Density 0.123%

    No Known Activations