© 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. 18861
    Prev
    Next
    INDEX
    Explanations

    scientific terms related to physical properties and phenomena

    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
    742
    +0.11
    0.5%
    677
    +0.10
    0.5%
    623
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    677
    +0.11
    0.04
    125
    +0.10
    0.04
    540
    +0.10
    0.04
    Negative Logits
    robe
    -0.68
    ombo
    -0.67
     humble
    -0.62
     sling
    -0.62
    STAR
    -0.57
    RIC
    -0.56
     resear
    -0.56
    odan
    -0.55
    OME
    -0.55
    ilings
    -0.54
    POSITIVE LOGITS
     everywhere
    1.07
     throughout
    0.89
     anywhere
    0.76
    iator
    0.74
     elsewhere
    0.74
     Everywhere
    0.73
     across
    0.71
     anytime
    0.71
     within
    0.71
    itored
    0.70
    Activations Density 0.170%

    No Known Activations