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. 23755
    Prev
    Next
    INDEX
    Explanations

    key attributes and features of a specific item or product

    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
    373
    +0.19
    0.9%
    545
    +0.12
    0.6%
    102
    +0.09
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    181
    +0.19
    0.04
    373
    +0.12
    -0.02
    227
    +0.09
    0.03
    Negative Logits
    ĸļ
    -0.69
    edom
    -0.68
     Adin
    -0.61
     Ri
    -0.60
    ctica
    -0.60
     Miche
    -0.60
    afety
    -0.57
    terday
    -0.56
    ngth
    -0.53
    idth
    -0.53
    POSITIVE LOGITS
    DragonMagazine
    0.75
    EStream
    0.68
    uls
    0.55
    CLASS
    0.54
    Spell
    0.54
    Dialogue
    0.54
    Introduced
    0.52
     smoked
    0.52
    cms
    0.51
     Logged
    0.51
    Activations Density 0.170%

    No Known Activations