© 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. 302
    Prev
    Next
    INDEX
    Explanations

    capital letters followed by 'o'

    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
    659
    +0.10
    0.5%
    670
    +0.10
    0.5%
    384
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.10
    0.02
    340
    +0.10
    0.02
    345
    +0.10
    0.02
    Negative Logits
     PID
    -0.75
     Yel
    -0.71
     Staples
    -0.70
    theless
    -0.70
    terday
    -0.69
     CoC
    -0.67
    channelAvailability
    -0.67
     Turing
    -0.67
     unfocusedRange
    -0.65
     FANT
    -0.64
    POSITIVE LOGITS
    phthal
    1.17
    atmeal
    1.15
    vernight
    1.12
    cean
    1.10
    asis
    1.09
    ops
    1.09
    lymp
    1.07
    mbudsman
    1.06
    ppo
    1.05
    culus
    1.04
    Activations Density 0.036%

    No Known Activations