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. 43183
    Prev
    Next
    INDEX
    Explanations

    references to phone calls or communication events

    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
    385
    +0.11
    0.5%
    481
    +0.10
    0.5%
    59
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    608
    +0.11
    0.03
    385
    +0.10
    0.02
    614
    +0.10
    0.02
    Negative Logits
    iencies
    -0.94
    estyles
    -0.84
    heet
    -0.77
    ival
    -0.76
    redits
    -0.74
    stones
    -0.73
    ikes
    -0.73
     pioneered
    -0.72
    selection
    -0.72
     stages
    -0.72
    POSITIVE LOGITS
     superiors
    0.86
     someone
    0.83
     voic
    0.82
     somebody
    0.75
     disgruntled
    0.74
     informing
    0.73
     whence
    0.71
     strangers
    0.70
     afar
    0.69
     Osama
    0.69
    Activations Density 0.043%

    No Known Activations