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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 12288
    5. 8-RES_FS12288-JB
    6. 9944
    Prev
    Next
    INDEX
    Explanations

    geographical locations and political events

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    640
    +0.16
    0.7%
    183
    +0.13
    0.6%
    91
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    183
    +0.16
    0.08
    91
    +0.13
    0.08
    257
    +0.12
    0.07
    Negative Logits
     unemploy
    -0.78
    reon
    -0.68
    $$
    -0.65
     combatants
    -0.65
     looms
    -0.64
    :/
    -0.64
     selves
    -0.63
     creeps
    -0.62
     RIGHT
    -0.61
     spoil
    -0.61
    POSITIVE LOGITS
     addition
    1.35
    cluding
    1.24
     spite
    1.22
     contrast
    1.22
    jured
    1.18
    ventory
    1.15
    spect
    1.15
    juries
    1.14
    flation
    1.13
    cluded
    1.10
    Activations Density 0.185%

    No Known Activations