© 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. 9309
    Prev
    Next
    INDEX
    Explanations

    details related to police incidents, shootings, criminal charges, and legal matters

    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
    481
    +0.21
    0.9%
    447
    +0.13
    0.6%
    146
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    481
    +0.21
    0.06
    146
    +0.13
    0.05
    15
    +0.12
    0.06
    Negative Logits
     "$:/
    -0.65
    anol
    -0.65
    ixel
    -0.64
    intel
    -0.62
    ngth
    -0.59
    regon
    -0.58
    dust
    -0.58
    asters
    -0.57
    enh
    -0.57
    erd
    -0.57
    POSITIVE LOGITS
     surfaced
    0.78
     concluded
    0.74
    >>>>>>>>
    0.74
     circulated
    0.72
     began
    0.70
    cens
    0.70
     vary
    0.69
     varies
    0.69
     discl
    0.69
     revealed
    0.68
    Activations Density 0.379%

    No Known Activations