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 3072
    5. 8-RES_FS3072-JB
    6. 1193
    Prev
    Next
    INDEX
    Explanations

    information related to news articles or reports

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    373
    +0.25
    1.2%
    530
    +0.11
    0.6%
    430
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    530
    +0.25
    0.10
    373
    +0.11
    -0.02
    764
    +0.09
    0.07
    Negative Logits
    username
    -0.65
    opter
    -0.58
    uds
    -0.58
    ttp
    -0.56
    orr
    -0.54
     Stall
    -0.54
    ilaterally
    -0.54
    tons
    -0.53
    href
    -0.53
    zos
    -0.53
    POSITIVE LOGITS
     CITY
    0.95
    WASHINGTON
    0.89
    SHARE
    0.80
    VILLE
    0.79
     COUNTY
    0.79
    LIN
    0.75
     JUL
    0.74
    MEN
    0.73
     MEN
    0.73
     SPR
    0.72
    Activations Density 0.722%

    No Known Activations