© 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 1536
    9. 8-RES_FS1536-JB
    11. 95
    Prev
    Next
    INDEX
    Explanations

    phrases related to potential involvement or participation

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    733
    +0.12
    0.5%
    125
    +0.10
    0.5%
    724
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    125
    +0.12
    0.10
    761
    +0.10
    0.10
    323
    +0.10
    0.09
    Negative Logits
    Pic
    -0.66
    STER
    -0.62
    iw
    -0.61
    catentry
    -0.61
    erves
    -0.61
    forge
    -0.61
    antry
    -0.60
    ESA
    -0.60
     <-
    -0.58
    gaard
    -0.58
    POSITIVE LOGITS
    tics
    0.72
     nicely
    0.66
    regor
    0.64
     roots
    0.62
     reinforcements
    0.61
    tml
    0.60
    escent
    0.60
    river
    0.58
     dissent
    0.58
     extra
    0.57
    Activations Density 2.132%

    No Known Activations