© 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. 402
    Prev
    Next
    INDEX
    Explanations

    words related to names and titles of people

    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
    97
    +0.13
    0.6%
    17
    +0.12
    0.5%
    267
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    99
    +0.13
    0.03
    291
    +0.12
    0.03
    97
    +0.11
    0.02
    Negative Logits
    ilater
    -0.99
    irlf
    -0.81
    itive
    -0.74
    ilogy
    -0.72
    ifully
    -0.69
    arians
    -0.69
    imilation
    -0.68
    ista
    -0.67
    eers
    -0.66
    éŃĶ
    -0.64
    POSITIVE LOGITS
    tics
    1.32
    rics
    1.15
    sis
    1.11
    puff
    1.04
    nda
    0.98
    mph
    0.95
    rical
    0.90
    phant
    0.89
    cemic
    0.88
    cos
    0.88
    Activations Density 0.105%

    No Known Activations