© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 49152
    9. 8-RES_FS49152-JB
    11. 26560
    Prev
    Next
    INDEX
    Explanations

    references to legislation and sponsorship in a political context

    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
    756
    +0.12
    0.6%
    350
    +0.11
    0.5%
    31
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    350
    +0.12
    0.02
    476
    +0.11
    0.01
    31
    +0.11
    0.02
    Negative Logits
    Pont
    -0.76
     Templ
    -0.75
     civilisation
    -0.72
     Leilan
    -0.66
    tics
    -0.65
     Jiu
    -0.63
     metab
    -0.61
     aber
    -0.60
     cis
    -0.60
     tsun
    -0.60
    POSITIVE LOGITS
    osponsors
    1.11
    ospons
    1.03
    sembly
    0.95
    eny
    0.93
    committee
    0.93
     motions
    0.92
     petitions
    0.92
    cong
    0.91
     amendments
    0.89
    clinton
    0.88
    Activations Density 0.030%

    No Known Activations