© 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. 321
    Prev
    Next
    INDEX
    Explanations

    the option to share or send content in various ways (email, social media, etc.) and information regarding promotions or giveaways

    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
    254
    +0.12
    0.5%
    481
    +0.11
    0.5%
    151
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.12
    0.08
    254
    +0.11
    0.06
    324
    +0.09
    0.05
    Negative Logits
    wed
    -0.65
    quickShipAvailable
    -0.64
    lander
    -0.64
    %%%%
    -0.63
     Theft
    -0.63
    ivalry
    -0.61
     manoeuv
    -0.60
    wrapper
    -0.58
    ideshow
    -0.58
    atech
    -0.57
    POSITIVE LOGITS
     coffers
    0.99
     via
    0.95
     detriment
    0.85
     ASAP
    0.81
     doorstep
    0.78
     shores
    0.74
     needy
    0.74
    omever
    0.73
     tune
    0.73
     unsuspecting
    0.72
    Activations Density 0.791%

    No Known Activations