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. 627
    Prev
    Next
    INDEX
    Explanations

    words related to critique and analysis of political and cultural topics

    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
    160
    +0.14
    0.6%
    314
    +0.13
    0.6%
    102
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    160
    +0.14
    0.05
    314
    +0.13
    0.01
    704
    +0.10
    0.02
    Negative Logits
     Bees
    -0.79
     tears
    -0.74
     bells
    -0.67
     oranges
    -0.67
     bows
    -0.67
     highlights
    -0.66
     Secrets
    -0.65
     Vaughn
    -0.64
     Inn
    -0.64
     Lynd
    -0.64
    POSITIVE LOGITS
    albeit
    1.51
    non
    1.30
    yet
    1.26
    sic
    1.25
    meaning
    1.16
    possibly
    1.10
    ?)
    1.06
    rather
    1.04
    adult
    1.04
    atomic
    1.03
    Activations Density 0.066%

    No Known Activations