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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 4-RES-SM
    11. 12416
    Prev
    Next
    INDEX
    Explanations

    words and phrases related to sexual assault and abuse

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__res-sm_processed/4-res-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.4.hook_resid_post
    Hook Layer
    4
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    186
    +0.14
    0.8%
    101
    +0.12
    0.7%
    62
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    186
    +0.14
    0.10
    101
    +0.12
    0.06
    189
    +0.11
    0.09
    Negative Logits
    č↵
    -4.05
    ↵
    -4.05
    ↵↵
    -4.05
    ↵
    -4.05
               ↵
    -4.05
    -4.05
    ↵↵
    -4.05
    ↵  ³³³
    -4.05
    -4.05
    ↵
    -4.05
    POSITIVE LOGITS
     victims
    2.91
     victim
    2.86
     punish
    2.72
     inflicted
    2.63
     perpet
    2.51
     suspect
    2.48
    ulent
    2.44
     prevention
    2.34
     tactics
    2.33
     against
    2.33
    Activations Density 1.168%

    No Known Activations