© 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 3072
    9. 8-RES_FS3072-JB
    11. 692
    Prev
    Next
    INDEX
    Explanations

    names of individuals, especially related to notable figures or positions

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    314
    +0.14
    0.6%
    373
    +0.14
    0.6%
    394
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    752
    +0.14
    0.12
    288
    +0.14
    0.13
    152
    +0.10
    0.10
    Negative Logits
     TAMADRA
    -0.74
    aturday
    -0.67
     iPads
    -0.65
     unnecessarily
    -0.63
     sufficiently
    -0.63
    Demand
    -0.62
    thora
    -0.61
    ynthesis
    -0.60
     Horus
    -0.60
     sockets
    -0.58
    POSITIVE LOGITS
     attorney
    1.22
    ologist
    1.20
     lieutenant
    1.18
     campaigner
    1.18
     director
    1.17
     adviser
    1.17
     researcher
    1.16
     strategist
    1.16
     staffer
    1.15
     lawyer
    1.15
    Activations Density 1.781%

    No Known Activations