© 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 12288
    9. 8-RES_FS12288-JB
    11. 772
    Prev
    Next
    INDEX
    Explanations

    the word "new" followed by a number

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    339
    +0.11
    0.5%
    296
    +0.10
    0.5%
    489
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    339
    +0.11
    0.06
    489
    +0.10
    0.06
    296
    +0.10
    0.06
    Negative Logits
    xual
    -0.79
    acebook
    -0.76
    istine
    -0.75
    OPA
    -0.70
    aminer
    -0.68
    osaurus
    -0.66
    itta
    -0.64
    ONSORED
    -0.64
    ILA
    -0.64
    UMP
    -0.64
    POSITIVE LOGITS
    bies
    1.54
    bie
    1.41
     arrivals
    1.32
     entrants
    1.15
    foundland
    1.14
     additions
    1.09
     Zealand
    1.07
     recruits
    0.99
     hires
    0.98
     developments
    0.97
    Activations Density 0.139%

    No Known Activations