© 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. 8042
    Prev
    Next
    INDEX
    Explanations

    sentences indicating beliefs or convictions

    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
    756
    +0.10
    0.5%
    597
    +0.10
    0.4%
    560
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    93
    +0.10
    0.02
    504
    +0.10
    0.02
    512
    +0.09
    0.02
    Negative Logits
    pin
    -0.76
     Summit
    -0.68
    IDER
    -0.68
     Palmer
    -0.67
    Interstitial
    -0.66
    PIN
    -0.65
     Gamma
    -0.65
    MIN
    -0.65
     Incarnation
    -0.64
     advis
    -0.64
    POSITIVE LOGITS
     they
    1.03
    they
    1.03
    she
    0.90
    erers
    0.86
    sbm
    0.81
     he
    0.81
    rained
    0.76
     we
    0.75
    ü
    0.73
    rists
    0.73
    Activations Density 0.107%

    No Known Activations