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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 1536
    5. 8-RES_FS1536-JB
    6. 591
    Prev
    Next
    INDEX
    Explanations

    phrases related to societal and economic issues

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    153
    +0.11
    0.5%
    676
    +0.11
    0.5%
    745
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    676
    +0.11
    0.11
    153
    +0.11
    0.10
    95
    +0.10
    0.07
    Negative Logits
    ratulations
    -0.73
    thia
    -0.61
    rounder
    -0.59
    ister
    -0.57
    LOG
    -0.56
    Reply
    -0.56
     transcripts
    -0.55
    sylv
    -0.54
    selling
    -0.54
    isters
    -0.53
    POSITIVE LOGITS
     behest
    1.41
     expense
    1.25
     outset
    1.22
     discretion
    1.06
     intersections
    1.04
     helm
    1.04
     forefront
    1.01
     doorstep
    0.95
     junction
    0.94
     mercy
    0.93
    Activations Density 1.552%

    No Known Activations