© 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 24576
    9. 8-RES_FS24576-JB
    11. 19850
    Prev
    Next
    INDEX
    Explanations

    phrases related to danger and risk to life

    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
    759
    +0.14
    0.6%
    81
    +0.10
    0.5%
    257
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    759
    +0.14
    0.04
    132
    +0.10
    0.04
    257
    +0.10
    0.02
    Negative Logits
    xual
    -0.76
     Compliance
    -0.74
    RECT
    -0.72
    ERC
    -0.70
     CLSID
    -0.68
     Rect
    -0.66
     franch
    -0.66
    uggest
    -0.65
    acco
    -0.64
    ansion
    -0.63
    POSITIVE LOGITS
    guards
    1.06
    mares
    0.97
    boats
    0.93
    ously
    0.86
    ffen
    0.86
    killers
    0.85
    vich
    0.85
     wounds
    0.84
     endanger
    0.84
     slain
    0.82
    Activations Density 0.053%

    No Known Activations