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 24576
    5. 8-RES_FS24576-JB
    6. 23626
    Prev
    Next
    INDEX
    Explanations

    references to ticket availability and purchasing for events

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    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
    415
    +0.13
    0.6%
    343
    +0.10
    0.5%
    705
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    415
    +0.13
    0.02
    370
    +0.10
    0.02
    426
    +0.10
    0.02
    Negative Logits
    lly
    -0.71
     advers
    -0.70
     Canaver
    -0.67
     investigative
    -0.66
     Cyr
    -0.66
     Forth
    -0.66
     Galile
    -0.66
     Schne
    -0.64
     motto
    -0.63
     invention
    -0.63
    POSITIVE LOGITS
     Tickets
    1.14
     tickets
    1.08
    Tickets
    1.04
    ticket
    0.90
     Ticket
    0.89
     holders
    0.86
     ticket
    0.81
    icket
    0.81
     booked
    0.78
     purchased
    0.76
    Activations Density 0.032%

    No Known Activations