Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 24576
    5. 8-RES_FS24576-JB
    6. 7768
    Prev
    Next
    INDEX
    Explanations

    terms related to electrical and computer engineering systems, as well as specific technologies and concepts within these fields

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    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
    +0.13
    0.6%
    687
    +0.11
    0.5%
    430
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.13
    0.05
    81
    +0.11
    0.04
    232
    +0.10
    0.04
    Negative Logits
    psons
    -0.94
     lockout
    -0.78
     tee
    -0.77
     preset
    -0.75
     axe
    -0.71
    isSpecialOrderable
    -0.71
     hose
    -0.70
     curfew
    -0.70
     blacklist
    -0.69
     badge
    -0.69
    POSITIVE LOGITS
     Studies
    1.28
     Sciences
    1.26
     Biology
    1.22
     Psychology
    1.21
     Physics
    1.16
    omics
    1.15
     Disorders
    1.13
    biology
    1.13
     Architecture
    1.12
     Science
    1.11
    Activations Density 0.205%

    No Known Activations