© 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 49152
    9. 8-RES_FS49152-JB
    11. 45529
    Prev
    Next
    INDEX
    Explanations

    references to body parts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    538
    +0.10
    0.5%
    532
    +0.10
    0.5%
    650
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    532
    +0.10
    0.03
    538
    +0.10
    0.02
    435
    +0.09
    0.02
    Negative Logits
    heny
    -0.62
    arenthood
    -0.62
    urdue
    -0.58
    versely
    -0.57
     Recomm
    -0.56
    qua
    -0.56
    theless
    -0.54
     interaction
    -0.54
     rarity
    -0.53
     hallucinations
    -0.53
    POSITIVE LOGITS
    plug
    0.70
    Plug
    0.70
     asses
    0.69
    Ĥ¬
    0.66
    ashington
    0.65
    ħĭ
    0.64
    horn
    0.63
    ache
    0.63
     burned
    0.62
     blown
    0.62
    Activations Density 0.057%

    No Known Activations