© 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. 7265
    Prev
    Next
    INDEX
    Explanations

    specific verbs relating to actions or decisions

    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
    +0.22
    1.0%
    4
    +0.10
    0.5%
    630
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    195
    +0.22
    0.03
    630
    +0.10
    0.05
    668
    +0.10
    0.04
    Negative Logits
    Versions
    -0.77
    selves
    -0.77
    Higher
    -0.76
    ordinary
    -0.71
    frequency
    -0.70
    Amount
    -0.68
    EMBER
    -0.67
    OTAL
    -0.67
    Basic
    -0.67
    ankind
    -0.66
    POSITIVE LOGITS
     adamant
    0.80
     herself
    0.77
     Tsarnaev
    0.74
     himself
    0.72
     scathing
    0.71
     coy
    0.70
     arra
    0.66
     emph
    0.65
     quoted
    0.65
     LLP
    0.65
    Activations Density 0.353%

    No Known Activations