© 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. 22798
    Prev
    Next
    INDEX
    Explanations

    words related to specific names or identities, particularly ones ending in 'ille'

    oai_token-act-pair · gpt-4o-miniTriggered 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
    710
    +0.14
    0.6%
    16
    +0.11
    0.5%
    119
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.14
    0.01
    99
    +0.11
    0.01
    110
    +0.10
    0.01
    Negative Logits
    iaries
    -0.78
     DPR
    -0.77
    -+-+
    -0.75
    etheless
    -0.75
    eanor
    -0.73
    reme
    -0.70
    iating
    -0.70
    anmar
    -0.68
    icum
    -0.68
    ortium
    -0.67
    POSITIVE LOGITS
    tto
    1.37
    tt
    1.22
    ttes
    1.20
    tta
    1.11
    hammer
    1.04
    lla
    1.01
    tti
    1.01
    brand
    0.99
    phant
    0.89
    mma
    0.88
    Activations Density 0.025%

    No Known Activations