© 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. 7174
    Prev
    Next
    INDEX
    Explanations

    references to recent events or statistics

    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
    339
    +0.10
    0.4%
    195
    +0.09
    0.4%
    146
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    622
    +0.10
    0.02
    339
    +0.09
    0.02
    253
    +0.08
    0.02
    Negative Logits
    claimer
    -0.74
    phy
    -0.72
    ocracy
    -0.72
    ldom
    -0.68
    iphate
    -0.68
    ocrine
    -0.67
    anguage
    -0.67
    onics
    -0.66
     ultimately
    -0.65
    igm
    -0.65
    POSITIVE LOGITS
     developments
    1.13
     trends
    1.10
     arrivals
    1.03
     generations
    1.01
     additions
    1.00
     events
    0.92
     controversies
    0.88
     revelations
    0.87
     observations
    0.86
     history
    0.84
    Activations Density 0.048%

    No Known Activations