© 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 EvalsExportsinterp-engineNEWAPI 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. 20843
    Prev
    Next
    INDEX
    Explanations

    information related to geographical locations and infrastructure details

    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
    89
    +0.10
    0.4%
    393
    +0.10
    0.4%
    635
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    89
    +0.10
    0.04
    765
    +0.10
    0.05
    43
    +0.09
    0.04
    Negative Logits
    irie
    -0.71
    ¤
    -0.71
    ²¾
    -0.70
    ESE
    -0.68
    icial
    -0.68
    ARCH
    -0.62
    ¥
    -0.60
    nesota
    -0.59
    '/
    -0.59
     è£ıè
    -0.58
    POSITIVE LOGITS
    000
    1.11
     whereas
    1.01
     respectively
    0.94
     according
    0.93
    Downloadha
    0.88
     although
    0.87
    depending
    0.85
    maximum
    0.85
     excluding
    0.85
     implying
    0.84
    Activations Density 0.165%

    No Known Activations