© 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 49152
    9. 8-RES_FS49152-JB
    11. 26525
    Prev
    Next
    INDEX
    Explanations

    phrases related to subjective opinion or interpretation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    373
    +0.11
    0.5%
    546
    +0.10
    0.4%
    598
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    642
    +0.11
    0.01
    535
    +0.10
    0.01
    547
    +0.09
    0.01
    Negative Logits
    incinn
    -0.73
    oute
    -0.70
    iland
    -0.70
    milo
    -0.69
    inate
    -0.69
    okane
    -0.67
    elta
    -0.66
    eni
    -0.64
    uala
    -0.63
    inav
    -0.62
    POSITIVE LOGITS
     aforementioned
    0.69
     usual
    0.67
     attendant
    0.67
     evidenced
    0.65
     envis
    0.65
    liest
    0.63
    same
    0.58
     posed
    0.58
     equation
    0.56
    ħĭ
    0.56
    Activations Density 0.010%

    No Known Activations