© 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. 10291
    Prev
    Next
    INDEX
    Explanations

    information related to political figures or events

    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
    679
    +0.17
    0.8%
    6
    +0.14
    0.6%
    640
    +0.13
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    6
    +0.17
    0.17
    679
    +0.14
    0.14
    440
    +0.13
    0.12
    Negative Logits
     ageing
    -0.83
     assass
    -0.75
     dips
    -0.75
     lifestyle
    -0.75
     immersive
    -0.75
     ingest
    -0.75
     talent
    -0.74
     regener
    -0.74
     abduct
    -0.74
     ecosystem
    -0.74
    POSITIVE LOGITS
     Nonetheless
    1.35
     Nevertheless
    1.35
     Instead
    1.33
     However
    1.33
     Afterwards
    1.30
     Then
    1.28
     Later
    1.26
     Eventually
    1.24
     Needless
    1.21
     Ultimately
    1.21
    Activations Density 0.790%

    No Known Activations