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. 41279
    Prev
    Next
    INDEX
    Explanations

    evaluative statements about actions or situations indicating their complexity or impact

    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
    87
    +0.27
    1.4%
    756
    +0.17
    0.8%
    459
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    756
    +0.27
    -0.00
    460
    +0.17
    0.04
    87
    +0.12
    -0.01
    Negative Logits
    ongevity
    -0.65
    inite
    -0.59
    venants
    -0.57
    oreal
    -0.56
    adium
    -0.55
    hens
    -0.53
    ioxide
    -0.53
    arnaev
    -0.52
    issions
    -0.52
    ensions
    -0.51
    POSITIVE LOGITS
    !).
    1.04
    !),
    0.84
    ).
    0.83
    ?).
    0.81
     anyway
    0.80
    !)
    0.78
     euphem
    0.75
    )}
    0.72
    )</
    0.72
     considering
    0.71
    Activations Density 0.344%

    No Known Activations