© 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 49152
    9. 8-RES_FS49152-JB
    11. 37328
    Prev
    Next
    INDEX
    Explanations

    proper nouns related to entities or individuals

    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
    447
    +0.13
    0.6%
    227
    +0.10
    0.5%
    391
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.13
    0.03
    232
    +0.10
    0.03
    643
    +0.10
    0.02
    Negative Logits
     psi
    -0.80
     VOL
    -0.74
     Shelley
    -0.74
     USPS
    -0.73
     KN
    -0.70
     stre
    -0.69
     Volvo
    -0.68
     MU
    -0.68
     Schl
    -0.67
     Phill
    -0.67
    POSITIVE LOGITS
    ad
    1.77
    ads
    1.55
    AD
    1.42
    adic
    1.34
    adh
    1.34
    adan
    1.22
    adian
    1.18
    ada
    1.17
    adin
    1.17
    ador
    1.17
    Activations Density 0.058%

    No Known Activations