© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 1368
    Prev
    Next
    INDEX
    Explanations

    words related to revealing information or uncovering secrets

    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
    101
    +0.11
    0.5%
    1
    +0.11
    0.5%
    288
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1
    +0.11
    0.03
    101
    +0.11
    0.03
    274
    +0.11
    0.03
    Negative Logits
    oslav
    -0.71
    otor
    -0.70
    stead
    -0.67
    hovah
    -0.67
    creation
    -0.66
    oday
    -0.66
     compensate
    -0.65
    acqu
    -0.64
    atomic
    -0.64
    upiter
    -0.64
    POSITIVE LOGITS
     secrets
    1.00
     loopholes
    0.98
     truths
    0.93
    ibility
    0.86
    orial
    0.83
     flaws
    0.81
     clues
    0.80
     weaknesses
    0.80
     details
    0.79
     revelations
    0.79
    Activations Density 0.067%

    No Known Activations