© 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 24576
    9. 8-RES_FS24576-JB
    11. 7193
    Prev
    Next
    INDEX
    Explanations

    information about different animals

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    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
    92
    +0.11
    0.5%
    762
    +0.10
    0.5%
    189
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    762
    +0.11
    0.06
    537
    +0.10
    0.04
    241
    +0.10
    0.04
    Negative Logits
    Fla
    -0.85
    Honest
    -0.76
    Recomm
    -0.70
    Ore
    -0.69
     Hills
    -0.67
    Tenn
    -0.67
    elight
    -0.66
    stown
    -0.65
    BUG
    -0.64
    Witness
    -0.64
    POSITIVE LOGITS
     are
    1.30
     themselves
    1.17
     aren
    1.12
     differ
    1.05
     were
    1.04
     belong
    1.03
     contain
    1.00
     constitute
    1.00
     behave
    0.99
     exist
    0.99
    Activations Density 0.239%

    No Known Activations