© 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. 18841
    Prev
    Next
    INDEX
    Explanations

    phrases related to resignation or stepping down from a position

    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
    447
    +0.11
    0.5%
    288
    +0.10
    0.5%
    488
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    199
    +0.11
    0.02
    37
    +0.10
    0.02
    488
    +0.08
    0.01
    Negative Logits
    âĨij
    -0.73
    gran
    -0.68
    Grab
    -0.64
    ONES
    -0.64
    Lex
    -0.63
    rouse
    -0.63
    creation
    -0.62
    rus
    -0.60
     Reprodu
    -0.59
    gat
    -0.59
    POSITIVE LOGITS
     resignation
    0.94
     resign
    0.94
     resigned
    0.94
    owship
    0.87
    uting
    0.87
     abruptly
    0.86
    utes
    0.86
    uted
    0.78
    iture
    0.76
    utive
    0.76
    Activations Density 0.020%

    No Known Activations