© 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. 37298
    Prev
    Next
    INDEX
    Explanations

    statements about existence, being, and the nature of self

    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
    480
    +0.23
    1.1%
    447
    +0.16
    0.8%
    450
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    762
    +0.23
    0.06
    566
    +0.16
    0.04
    718
    +0.10
    0.04
    Negative Logits
     GoPro
    -0.84
    multiple
    -0.81
     trending
    -0.76
    ICO
    -0.75
     CTR
    -0.75
    impact
    -0.74
    boost
    -0.74
     Crunch
    -0.73
    Ops
    -0.72
     marquee
    -0.72
    POSITIVE LOGITS
     confess
    1.06
     pity
    1.06
     consec
    1.05
     pious
    1.05
     vain
    1.01
     theolog
    1.01
     philosophers
    0.98
     divine
    0.97
     blasp
    0.97
     scarcely
    0.96
    Activations Density 0.413%

    No Known Activations