© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 9500
    Prev
    Next
    INDEX
    Explanations

    and highlight instances of abuse or inappropriate behavior in a document

    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
    481
    +0.13
    0.6%
    299
    +0.11
    0.5%
    447
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    393
    +0.13
    0.07
    328
    +0.11
    0.05
    720
    +0.10
    0.06
    Negative Logits
    baum
    -0.86
    gue
    -0.82
    ciples
    -0.80
    eric
    -0.78
    eport
    -0.78
    heast
    -0.76
    arten
    -0.76
    ulum
    -0.76
    utical
    -0.76
    cade
    -0.74
    POSITIVE LOGITS
     degradation
    1.25
     persecution
    1.21
     vandalism
    1.20
     harassment
    1.18
     suicides
    1.18
     criminality
    1.17
     theft
    1.15
     illness
    1.14
     thefts
    1.14
     mayhem
    1.12
    Activations Density 0.558%

    No Known Activations