© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 8779
    Prev
    Next
    INDEX
    Explanations

    words related to a specific entity or person named "Ver"

    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
    545
    +0.13
    0.6%
    631
    +0.11
    0.5%
    727
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    545
    +0.13
    0.02
    631
    +0.11
    0.02
    702
    +0.10
    0.02
    Negative Logits
    ħĭ
    -0.73
     Sparrow
    -0.73
    »Ĵ
    -0.68
    ollo
    -0.67
    urst
    -0.66
    ¬¼
    -0.64
    ij士
    -0.63
     listen
    -0.62
     havens
    -0.61
     needle
    -0.61
    POSITIVE LOGITS
    dict
    1.27
    tex
    1.06
    gil
    1.03
    million
    1.03
    onica
    1.01
    ifying
    0.99
    itas
    0.99
    bal
    0.97
    gent
    0.96
    ified
    0.95
    Activations Density 0.012%

    No Known Activations