© 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
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 24576
    5. 8-RES_FS24576-JB
    6. 7291
    Prev
    Next
    INDEX
    Explanations

    the conjunction "and"

    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
    209
    +0.09
    0.4%
    648
    +0.08
    0.4%
    585
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    284
    +0.09
    0.03
    746
    +0.08
    0.02
    484
    +0.08
    0.03
    Negative Logits
    Magikarp
    -0.63
    âĢ¢âĢ¢
    -0.60
    Ly
    -0.54
     GPS
    -0.54
     Kamp
    -0.52
     necks
    -0.52
     RPGs
    -0.51
    Charges
    -0.51
    OGR
    -0.50
     Dunk
    -0.50
    POSITIVE LOGITS
    rogen
    1.08
    rogens
    0.95
    then
    0.79
    thens
    0.73
    ¬¼
    0.71
    rology
    0.70
    every
    0.68
     sund
    0.68
    rew
    0.67
    aspx
    0.66
    Activations Density 0.083%

    No Known Activations