© 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 24576
    9. 8-RES_FS24576-JB
    11. 1722
    Prev
    Next
    INDEX
    Explanations

    details related to different locations and measurements, such as sites, cells, and directions

    oai_token-act-pair · gpt-3.5-turboTriggered 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
    480
    +0.46
    2.3%
    716
    +0.09
    0.4%
    202
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.46
    0.10
    480
    +0.09
    0.00
    266
    +0.08
    0.08
    Negative Logits
     âĵĺ
    -0.66
    terday
    -0.66
    ovember
    -0.63
     Rica
    -0.59
     Leone
    -0.58
    olia
    -0.57
    iatus
    -0.56
    utics
    -0.56
    DonaldTrump
    -0.56
     Month
    -0.54
    POSITIVE LOGITS
    iest
    1.01
     thereof
    0.96
     equivalent
    0.95
    liest
    0.94
     portion
    0.90
     itself
    0.89
    osphere
    0.87
     ones
    0.83
     equivalents
    0.81
     nearest
    0.79
    Activations Density 0.789%

    No Known Activations