© 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 24576
    9. 8-RES_FS24576-JB
    11. 15676
    Prev
    Next
    INDEX
    Explanations

    social media-related words and terms, with a particular focus on the concept of followers

    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
    679
    +0.13
    0.6%
    105
    +0.10
    0.5%
    469
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    32
    +0.13
    0.01
    469
    +0.10
    0.01
    51
    +0.10
    0.01
    Negative Logits
    ces
    -0.72
     Rim
    -0.65
     Genocide
    -0.64
    senal
    -0.64
     Barnett
    -0.63
    posing
    -0.62
    cer
    -0.62
     Prosecutor
    -0.61
    ced
    -0.61
    geon
    -0.59
    POSITIVE LOGITS
    hip
    1.45
    hips
    1.10
     followers
    1.00
    lihood
    0.85
     follower
    0.81
    wagon
    0.73
    ollower
    0.73
     wagon
    0.72
    fulness
    0.72
     gladly
    0.71
    Activations Density 0.039%

    No Known Activations