© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 12288
    9. 8-RES_FS12288-JB
    11. 310
    Prev
    Next
    INDEX
    Explanations

    phrases prompting interaction or sharing of opinions

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    715
    +0.13
    0.6%
    608
    +0.10
    0.5%
    234
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    608
    +0.13
    0.06
    665
    +0.10
    0.06
    715
    +0.10
    0.03
    Negative Logits
    lehem
    -0.75
    UF
    -0.69
    awei
    -0.65
     millenn
    -0.65
    alky
    -0.64
    Dur
    -0.63
    bush
    -0.62
    lines
    -0.61
     Compass
    -0.61
    urance
    -0.60
    POSITIVE LOGITS
     indulge
    0.91
     speculate
    0.87
     criticize
    0.85
     disagree
    0.84
     explore
    0.82
     improv
    0.81
     roam
    0.79
     assume
    0.78
     discuss
    0.76
     conjecture
    0.75
    Activations Density 0.252%

    No Known Activations