© 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 3072
    9. 8-RES_FS3072-JB
    11. 2959
    Prev
    Next
    INDEX
    Explanations

    technology-related terms and product/service names

    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
    +0.12
    0.6%
    623
    +0.10
    0.5%
    176
    +0.09
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.12
    0.15
    659
    +0.10
    0.14
    314
    +0.09
    0.00
    Negative Logits
     cryptocurrency
    -0.57
     bitcoin
    -0.57
     compared
    -0.54
     worthwhile
    -0.54
     seriously
    -0.53
     sufficiently
    -0.53
     depl
    -0.52
     rather
    -0.52
     whistleblower
    -0.52
     theirs
    -0.52
    POSITIVE LOGITS
    cellaneous
    0.72
    ikuman
    0.66
    oultry
    0.65
    odka
    0.62
    rador
    0.60
    zes
    0.59
    agog
    0.58
    ĥ
    0.58
     etc
    0.57
    uania
    0.57
    Activations Density 2.649%

    No Known Activations