© 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 3072
    9. 8-RES_FS3072-JB
    11. 3036
    Prev
    Next
    INDEX
    Explanations

    numeric information like quantities, measurements, or rankings

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    93
    +0.11
    0.5%
    678
    +0.09
    0.4%
    0
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    409
    +0.11
    0.08
    278
    +0.09
    0.08
    81
    +0.09
    0.08
    Negative Logits
    Ô
    -0.89
    Rh
    -0.81
    nos
    -0.79
    Northern
    -0.79
    Yan
    -0.79
    Southern
    -0.77
    ighed
    -0.77
    Wil
    -0.76
    Æ
    -0.75
    Haw
    -0.75
    POSITIVE LOGITS
     panel
    0.99
     system
    0.98
     scheme
    0.92
     unit
    0.91
     chamber
    0.90
     kit
    0.89
     cohort
    0.88
     device
    0.87
     platform
    0.87
     portfolio
    0.86
    Activations Density 1.795%

    No Known Activations