© 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. 1615
    Prev
    Next
    INDEX
    Explanations

    phrases related to actions taken on people or objects

    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
    33
    +0.10
    0.5%
    81
    +0.09
    0.4%
    533
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    665
    +0.10
    0.08
    70
    +0.09
    0.08
    715
    +0.09
    0.06
    Negative Logits
     ][
    -0.75
    ocument
    -0.65
    ortium
    -0.63
    ague
    -0.61
    ibrary
    -0.58
     and
    -0.58
    ocaust
    -0.57
    uay
    -0.57
    enny
    -0.56
    berman
    -0.56
    POSITIVE LOGITS
     accordingly
    1.12
     them
    0.96
     alike
    0.89
     THEM
    0.77
     consequ
    0.73
     thereafter
    0.73
     consequently
    0.69
     others
    0.68
     ourselves
    0.68
     him
    0.68
    Activations Density 0.766%

    No Known Activations