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 49152
    5. 8-RES_FS49152-JB
    6. 26008
    Prev
    Next
    INDEX
    Explanations

    terms related to measures, actions, or considerations taken for specific purposes or effects

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    481
    +0.24
    1.1%
    483
    +0.10
    0.5%
    460
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    460
    +0.24
    0.04
    312
    +0.10
    0.03
    221
    +0.10
    0.03
    Negative Logits
    arching
    -0.65
    indal
    -0.60
    jing
    -0.58
    oreal
    -0.57
     tremend
    -0.57
    ourses
    -0.56
    qqa
    -0.55
    risome
    -0.54
    afety
    -0.53
    ufact
    -0.53
    POSITIVE LOGITS
    ,
    0.77
    ,.
    0.74
    ,,
    0.67
     there
    0.65
    *,
    0.62
     we
    0.61
     congr
    0.61
    Ø©
    0.59
    :
    0.58
    )(
    0.56
    Activations Density 0.110%

    No Known Activations