© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 10294
    Prev
    Next
    INDEX
    Explanations

    ages or numeric values indicating a person's age

    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
    762
    +0.12
    0.5%
    72
    +0.12
    0.5%
    208
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    72
    +0.12
    0.03
    762
    +0.12
    0.02
    545
    +0.11
    0.02
    Negative Logits
     coord
    -0.66
     periphery
    -0.60
     subp
    -0.59
    tap
    -0.57
    illance
    -0.57
     mathemat
    -0.56
    vernment
    -0.55
    SN
    -0.55
     sidel
    -0.55
    Ts
    -0.55
    POSITIVE LOGITS
    years
    0.80
    enough
    0.78
     considering
    0.71
     (~
    0.70
    ÑĤ
    0.67
    !--
    0.67
    digy
    0.66
    stable
    0.65
     enough
    0.65
    debian
    0.65
    Activations Density 0.050%

    No Known Activations