© 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. 9701
    Prev
    Next
    INDEX
    Explanations

    names of people or companies with a focus on financial or technological contexts

    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
    294
    +0.12
    0.5%
    224
    +0.11
    0.5%
    6
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    6
    +0.12
    0.05
    176
    +0.11
    0.03
    294
    +0.11
    0.03
    Negative Logits
    cephal
    -0.86
    aily
    -0.85
    achev
    -0.81
    unal
    -0.79
    imet
    -0.78
    jriwal
    -0.77
    ciplinary
    -0.75
    bole
    -0.75
    obal
    -0.72
    emetery
    -0.70
    POSITIVE LOGITS
     SG
    1.36
     PG
    1.30
     LD
    1.30
     MI
    1.29
     MN
    1.28
     NT
    1.28
     HI
    1.27
     TX
    1.25
     WS
    1.25
     IL
    1.25
    Activations Density 0.336%

    No Known Activations