Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 49152
    5. 8-RES_FS49152-JB
    6. 26130
    Prev
    Next
    INDEX
    Explanations

    terms related to military and national security issues

    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
    481
    +0.22
    1.0%
    362
    +0.11
    0.5%
    713
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.22
    0.06
    67
    +0.11
    0.04
    362
    +0.09
    0.04
    Negative Logits
    ensor
    -0.68
    slot
    -0.68
    */(
    -0.66
    utterstock
    -0.66
    gue
    -0.65
     Examiner
    -0.64
    handle
    -0.64
    izons
    -0.63
    ainer
    -0.62
    icator
    -0.62
    POSITIVE LOGITS
     wrongdoing
    1.08
     discrepancies
    1.07
    Semitism
    1.07
     irregularities
    1.05
     atrocities
    1.03
     malf
    0.97
     violations
    0.95
     injust
    0.93
     abuses
    0.93
     suicides
    0.89
    Activations Density 0.266%

    No Known Activations