© 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. 10103
    Prev
    Next
    INDEX
    Explanations

    specific information related to events or appearances of individuals

    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
    362
    +0.15
    0.7%
    393
    +0.15
    0.7%
    308
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.15
    0.07
    16
    +0.15
    0.05
    327
    +0.11
    0.04
    Negative Logits
    Query
    -0.72
    acon
    -0.67
    gradient
    -0.66
    BIL
    -0.66
    uated
    -0.65
    ère
    -0.64
    isson
    -0.64
    washed
    -0.61
    588
    -0.60
    _-
    -0.60
    POSITIVE LOGITS
     world
    1.25
     entire
    1.08
     shortest
    1.07
     country
    1.02
     nation
    1.01
     universe
    1.00
     vicinity
    0.99
     history
    0.97
     midst
    0.94
     entirety
    0.92
    Activations Density 0.169%

    No Known Activations