Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. Residuals
    5. 2-RES-SM
    6. 29012
    Prev
    Next
    INDEX
    Explanations

    terms and phrases related to military, veterans, and their experiences

    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/2-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.2.hook_resid_post
    Hook Layer
    2
    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
    156
    +0.24
    1.4%
    186
    +0.13
    0.7%
    304
    +0.11
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    418
    +0.24
    0.00
    186
    +0.13
    0.02
    304
    +0.11
    0.03
    Negative Logits
    opic
    -1.92
    woke
    -1.89
     dreamed
    -1.65
     coli
    -1.64
     someone
    -1.63
     Freud
    -1.63
    rogens
    -1.63
    TRODUCTION
    -1.61
     yours
    -1.60
     chromosomal
    -1.59
    POSITIVE LOGITS
    ĸ
    2.08
    ²
    1.87
    ¥
    1.86
    ¡
    1.82
     officer
    1.75
     uniforms
    1.75
     Force
    1.70
    fare
    1.67
     contractor
    1.66
    Ī
    1.60
    Activations Density 0.250%

    No Known Activations