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. 23895
    Prev
    Next
    INDEX
    Explanations

    references to size or the concept of "bigger"

    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
    623
    +0.13
    0.6%
    487
    +0.10
    0.5%
    654
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    623
    +0.13
    0.03
    65
    +0.10
    0.02
    487
    +0.10
    0.02
    Negative Logits
    shire
    -0.76
    EVA
    -0.73
    pta
    -0.72
    syn
    -0.69
     Immun
    -0.67
    hire
    -0.67
    Grade
    -0.64
    odor
    -0.64
    WP
    -0.64
    EP
    -0.64
    POSITIVE LOGITS
     than
    1.16
    picture
    1.08
    than
    1.00
     picture
    0.96
     Than
    0.94
    oted
    0.82
     scale
    0.80
     fish
    0.77
     bang
    0.75
     chunks
    0.74
    Activations Density 0.019%

    No Known Activations