© 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. 37221
    Prev
    Next
    INDEX
    Explanations

    references to various grades or levels of excellence

    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
    352
    +0.11
    0.5%
    357
    +0.10
    0.5%
    437
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    357
    +0.11
    0.03
    352
    +0.10
    0.03
    437
    +0.10
    0.03
    Negative Logits
     Forth
    -0.70
     questioning
    -0.63
     Remem
    -0.63
     Nun
    -0.63
     reclaimed
    -0.62
    terday
    -0.62
     constitu
    -0.60
     Isles
    -0.60
    GROUND
    -0.60
     Dear
    -0.59
    POSITIVE LOGITS
    etermin
    1.24
    ummies
    1.21
    ivid
    1.20
    ildo
    1.18
    renched
    1.18
    ividual
    1.16
    itto
    1.16
    ashing
    1.16
    ashes
    1.13
    azed
    1.12
    Activations Density 0.016%

    No Known Activations