© 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. 7276
    Prev
    Next
    INDEX
    Explanations

    words related to political or social commentary with negative connotations

    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
    232
    +0.12
    0.5%
    103
    +0.10
    0.4%
    383
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.12
    0.02
    103
    +0.10
    0.02
    137
    +0.09
    0.01
    Negative Logits
    Construct
    -0.84
    geist
    -0.72
    UAL
    -0.70
    RAL
    -0.70
    AGE
    -0.70
    ETF
    -0.67
    XY
    -0.65
    reon
    -0.65
    lectic
    -0.64
     Gould
    -0.64
    POSITIVE LOGITS
    ossom
    1.27
    izzard
    1.16
    ameless
    1.13
    oop
    1.07
    adders
    1.05
    itzer
    1.05
    anche
    1.05
    ogging
    1.01
    anks
    1.00
    urring
    1.00
    Activations Density 0.012%

    No Known Activations