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. 21434
    Prev
    Next
    INDEX
    Explanations

    C++ code syntax related to pointers and strings within a larger text context

    oai_token-act-pair · gpt-3.5-turboTriggered 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
    480
    +0.43
    2.4%
    393
    +0.11
    0.6%
    154
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.43
    0.04
    667
    +0.11
    0.03
    28
    +0.10
    0.04
    Negative Logits
    osponsors
    -0.51
     EXP
    -0.48
    compan
    -0.45
     Insp
    -0.45
     supervised
    -0.44
     calendars
    -0.44
     sponsors
    -0.44
     organized
    -0.44
     organised
    -0.43
     November
    -0.43
    POSITIVE LOGITS
    obal
    0.58
    tein
    0.57
    cffffcc
    0.56
    /+
    0.56
    thora
    0.55
    govtrack
    0.54
    rican
    0.54
    DonaldTrump
    0.54
    ntil
    0.53
     vomit
    0.52
    Activations Density 0.479%

    No Known Activations