© 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 49152
    9. 8-RES_FS49152-JB
    11. 45412
    Prev
    Next
    INDEX
    Explanations

    instances of the word "h" and variations with different activation values

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    240
    +0.12
    0.6%
    373
    +0.12
    0.6%
    134
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.12
    0.02
    701
    +0.12
    0.01
    332
    +0.11
    0.02
    Negative Logits
     Eternity
    -0.71
     mosqu
    -0.65
     ducks
    -0.64
     Bots
    -0.64
     conclud
    -0.64
     destro
    -0.61
    eering
    -0.61
     wedge
    -0.60
     heartbeat
    -0.60
     Clicker
    -0.60
    POSITIVE LOGITS
    ulhu
    1.24
    orses
    1.10
    arma
    1.05
    atever
    1.03
    agen
    1.03
    arel
    1.03
    ilar
    1.02
    ollow
    0.97
    airy
    0.95
    onest
    0.94
    Activations Density 0.011%

    No Known Activations