Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 6144
    5. 8-RES_FS6144-JB
    6. 4551
    Prev
    Next
    INDEX
    Explanations

    the phrase "time and again."

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs6144-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_6144
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    6,144
    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
    136
    +0.13
    0.6%
    373
    +0.12
    0.5%
    710
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    136
    +0.13
    0.06
    546
    +0.12
    0.04
    278
    +0.10
    0.04
    Negative Logits
     Cosponsors
    -0.79
    fml
    -0.74
     Marketable
    -0.71
    -+-+
    -0.68
    Wan
    -0.67
    DNA
    -0.66
     Supports
    -0.65
     Survive
    -0.65
    è¦ļéĨĴ
    -0.63
     rall
    -0.63
    POSITIVE LOGITS
     lapse
    0.75
    vity
    0.71
     wasted
    0.70
     expended
    0.67
    theless
    0.67
    times
    0.67
    ftime
    0.67
    today
    0.66
    obos
    0.65
     stress
    0.63
    Activations Density 0.622%

    No Known Activations