© 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. 10256
    Prev
    Next
    INDEX
    Explanations

    phrases related to news and information

    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
    373
    +0.59
    5.0%
    481
    +0.51
    4.3%
    480
    +0.27
    2.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    373
    +0.59
    -0.10
    481
    +0.51
    0.16
    480
    +0.27
    -0.04
    Negative Logits
    emale
    -0.36
    romy
    -0.34
    omach
    -0.33
    nesday
    -0.32
    uesday
    -0.32
    Firstly
    -0.32
    dfx
    -0.32
    adra
    -0.31
    ³³³³³³³³³³³³³³³³
    -0.31
    0002
    -0.31
    POSITIVE LOGITS
    .'"
    0.30
     Huck
    0.30
    —"
    0.30
     nonetheless
    0.29
    ))))
    0.29
     aggrav
    0.28
     Gib
    0.28
    '."
    0.28
     HuffPost
    0.28
     Saban
    0.27
    Activations Density 3.220%

    No Known Activations