© 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 12288
    9. 8-RES_FS12288-JB
    11. 1179
    Prev
    Next
    INDEX
    Explanations

    historical events and figures

    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
    168
    +0.15
    0.7%
    728
    +0.12
    0.6%
    659
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.15
    0.08
    232
    +0.12
    0.07
    141
    +0.12
    0.05
    Negative Logits
    Safety
    -0.82
     opioids
    -0.78
    FOX
    -0.78
     scanner
    -0.76
     trending
    -0.76
     clip
    -0.76
     fluorescent
    -0.75
     inputs
    -0.75
    tools
    -0.75
     cannabinoids
    -0.74
    POSITIVE LOGITS
     Augustus
    1.37
     XVI
    1.30
     VIII
    1.26
     XIV
    1.22
     Tud
    1.20
     Herod
    1.18
     XII
    1.13
     Ferdinand
    1.13
     Claud
    1.12
     VII
    1.10
    Activations Density 0.485%

    No Known Activations