© 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. 45819
    Prev
    Next
    INDEX
    Explanations

    references to identities and names

    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
    592
    +0.14
    0.6%
    28
    +0.13
    0.6%
    278
    +0.13
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    28
    +0.14
    0.02
    278
    +0.13
    0.01
    461
    +0.13
    0.01
    Negative Logits
    HER
    -0.85
    710
    -0.83
    oka
    -0.82
    570
    -0.80
    ibilities
    -0.79
    615
    -0.78
    540
    -0.77
    420
    -0.76
    ohm
    -0.75
    550
    -0.75
    POSITIVE LOGITS
    de
    1.08
     de
    1.07
     De
    0.99
     des
    0.94
    des
    0.90
    De
    0.88
     Des
    0.87
    DE
    0.86
     DE
    0.82
     Dele
    0.82
    Activations Density 0.042%

    No Known Activations