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. 12121
    Prev
    Next
    INDEX
    Explanations

    instances of the term "event" used in various contexts related to occurrences or situations

    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
    408
    +0.12
    0.6%
    535
    +0.11
    0.5%
    496
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    535
    +0.12
    0.01
    599
    +0.11
    0.01
    408
    +0.10
    0.01
    Negative Logits
     Flavoring
    -0.81
    pees
    -0.75
    elta
    -0.74
    clud
    -0.73
    arag
    -0.70
    PAC
    -0.70
    cest
    -0.69
    uph
    -0.67
    ilings
    -0.65
     Topic
    -0.65
    POSITIVE LOGITS
     circumstances
    0.79
     thereof
    0.76
     they
    0.70
     unsuccessful
    0.69
     unsatisf
    0.65
     lapse
    0.65
     euth
    0.65
    eers
    0.64
     complications
    0.63
     we
    0.63
    Activations Density 0.009%

    No Known Activations