© 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. 19712
    Prev
    Next
    INDEX
    Explanations

    names of individuals

    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
    356
    +0.11
    0.5%
    145
    +0.10
    0.4%
    660
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    701
    +0.11
    0.02
    145
    +0.10
    0.02
    299
    +0.09
    0.02
    Negative Logits
    glers
    -0.83
     hail
    -0.68
     cones
    -0.62
     bother
    -0.58
    fires
    -0.58
     fork
    -0.58
     cigars
    -0.57
     spiders
    -0.57
     Typhoon
    -0.57
     snakes
    -0.56
    POSITIVE LOGITS
    illary
    1.00
    wered
    1.00
    heim
    0.94
    imity
    0.93
    otropic
    0.88
    obic
    0.83
    ibo
    0.82
    orage
    0.81
    andre
    0.79
    estic
    0.76
    Activations Density 0.041%

    No Known Activations