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. 23816
    Prev
    Next
    INDEX
    Explanations

    phrases and expressions indicating timeframes or events in a sequence

    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
    703
    +0.12
    0.5%
    346
    +0.10
    0.5%
    410
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    703
    +0.12
    0.04
    98
    +0.10
    0.05
    172
    +0.09
    0.04
    Negative Logits
    acci
    -0.74
    Peace
    -0.73
     constitu
    -0.71
    abus
    -0.71
    enf
    -0.70
    pmwiki
    -0.69
     endif
    -0.68
    et
    -0.67
    IZE
    -0.67
    MENT
    -0.66
    POSITIVE LOGITS
     undergoing
    0.97
     retiring
    0.96
     finishing
    0.95
     acquiring
    0.94
     losing
    0.94
     completing
    0.92
     stint
    0.91
     joining
    0.87
     dominating
    0.87
    wards
    0.86
    Activations Density 0.129%

    No Known Activations