© 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. 9811
    Prev
    Next
    INDEX
    Explanations

    terms related to computer programming and code

    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
    288
    +0.13
    0.6%
    515
    +0.10
    0.5%
    457
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.13
    0.04
    326
    +0.10
    0.03
    515
    +0.10
    0.02
    Negative Logits
    akening
    -0.88
    yer
    -0.82
    hari
    -0.79
    uph
    -0.78
    spring
    -0.73
    ered
    -0.72
    fulness
    -0.72
    ski
    -0.69
    side
    -0.69
    sheet
    -0.67
    POSITIVE LOGITS
     emulator
    1.02
    ysis
    0.69
     terminal
    0.68
     mosqu
    0.67
    POS
    0.67
    xes
    0.67
    igmatic
    0.66
    TY
    0.65
    taining
    0.64
    fo
    0.63
    Activations Density 0.158%

    No Known Activations