© 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 768
    9. 8-RES_FS768-JB
    11. 690
    Prev
    Next
    INDEX
    Explanations

    embedded videos and online footage

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs768-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_768
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    768
    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
    679
    +0.12
    0.5%
    358
    +0.11
    0.5%
    475
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    358
    +0.12
    0.10
    498
    +0.11
    0.14
    679
    +0.09
    0.21
    Negative Logits
     Flavoring
    -0.76
    apesh
    -0.73
    ebus
    -0.73
     retaliate
    -0.72
    joice
    -0.66
    aido
    -0.65
     mobility
    -0.64
    idable
    -0.64
    BILITIES
    -0.64
    mare
    -0.64
    POSITIVE LOGITS
     excerpts
    1.45
     pdf
    1.34
     titled
    1.32
     excerpt
    1.27
     reprinted
    1.24
     archived
    1.21
     edited
    1.20
     reproduced
    1.20
     archive
    1.18
     redacted
    1.18
    Activations Density 9.280%

    No Known Activations