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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 1278
    Prev
    Next
    INDEX
    Explanations

    the verb "do" followed by a number indicating its activation strength

    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
    17
    +0.12
    0.5%
    78
    +0.11
    0.5%
    611
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    78
    +0.12
    0.07
    17
    +0.11
    0.06
    611
    +0.10
    0.06
    Negative Logits
    lights
    -0.73
    pac
    -0.61
    dayName
    -0.61
    mares
    -0.60
    sg
    -0.59
    ware
    -0.59
    DOC
    -0.59
    opol
    -0.58
    uses
    -0.57
     Prosecutor
    -0.56
    POSITIVE LOGITS
    omsday
    1.01
    ppel
    0.94
    pez
    0.91
    omething
    0.90
    lez
    0.85
    xx
    0.83
     something
    0.82
     laundry
    0.82
    ggy
    0.81
    oms
    0.80
    Activations Density 0.170%

    No Known Activations