© 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. 26842
    Prev
    Next
    INDEX
    Explanations

    repetitions and emphasis on the word "so."

    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
    55
    +0.12
    0.5%
    567
    +0.11
    0.5%
    320
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    567
    +0.12
    0.04
    55
    +0.11
    0.04
    11
    +0.10
    0.03
    Negative Logits
     Nex
    -0.76
    works
    -0.66
    stadt
    -0.62
     Faul
    -0.60
     burner
    -0.60
    amac
    -0.59
     Landing
    -0.59
     Reynolds
    -0.59
    backs
    -0.58
    eties
    -0.58
    POSITIVE LOGITS
    iled
    1.08
    apy
    1.03
    bered
    0.98
    oooo
    0.94
    iling
    0.93
     far
    0.91
    othing
    0.90
    oths
    0.88
     thoroughly
    0.85
     much
    0.84
    Activations Density 0.038%

    No Known Activations