© 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. 20590
    Prev
    Next
    INDEX
    Explanations

    instructions or technical details related to applications on phones or computers

    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
    430
    +0.11
    0.5%
    699
    +0.10
    0.5%
    214
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    655
    +0.11
    0.04
    699
    +0.10
    0.03
    214
    +0.10
    0.03
    Negative Logits
    onder
    -0.69
     Dynamics
    -0.68
    NEY
    -0.63
    orers
    -0.61
    roller
    -0.59
    taboola
    -0.59
    neys
    -0.58
    olitan
    -0.58
    rend
    -0.57
    ãĤ«
    -0.57
    POSITIVE LOGITS
     usual
    0.80
    normal
    0.78
    capacity
    0.75
    onyms
    0.74
    daq
    0.72
    average
    0.71
     5000
    0.70
     half
    0.69
    idth
    0.67
     oneself
    0.66
    Activations Density 0.074%

    No Known Activations