© 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. 399
    Prev
    Next
    INDEX
    Explanations

    words related to strong opinions or judgments

    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
    99
    +0.10
    0.5%
    284
    +0.10
    0.5%
    588
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    284
    +0.10
    0.03
    588
    +0.10
    0.03
    767
    +0.09
    0.02
    Negative Logits
    ĸļ
    -0.80
     warr
    -0.72
    DragonMagazine
    -0.68
     Parenthood
    -0.63
     ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³
    -0.63
    à©
    -0.61
     Tanz
    -0.61
     Grizz
    -0.61
    eering
    -0.60
    ãĤ°
    -0.60
    POSITIVE LOGITS
    chio
    1.19
    otine
    1.06
    ulously
    1.04
    ulous
    1.03
    orp
    1.01
    ourt
    1.00
    hes
    1.00
    henko
    0.99
    ulum
    0.98
    entric
    0.98
    Activations Density 0.075%

    No Known Activations