© 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. 11033
    Prev
    Next
    INDEX
    Explanations

    words related to work, pay, and possibly gender discrimination

    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
    127
    +0.12
    0.6%
    735
    +0.10
    0.5%
    264
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    283
    +0.12
    0.02
    591
    +0.10
    0.04
    212
    +0.10
    0.04
    Negative Logits
    enance
    -0.74
    luence
    -0.68
    ablished
    -0.63
    equality
    -0.62
    resents
    -0.61
    CONCLUS
    -0.60
    ablish
    -0.60
    Failure
    -0.59
    edience
    -0.58
    ifference
    -0.58
    POSITIVE LOGITS
    !).
    1.52
    ?).
    1.47
    !),
    1.41
    !)
    1.34
    ?),
    1.33
    ).
    1.25
    *)
    1.24
    >)
    1.23
    ?)
    1.21
    )).
    1.19
    Activations Density 0.615%

    No Known Activations