© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 2-RES-SM
    11. 27360
    Prev
    Next
    INDEX
    Explanations

    references to national identity or national issues

    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
    +0.23
    1.3%
    30
    +0.18
    1.1%
    405
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    30
    +0.23
    0.03
    119
    +0.18
    0.01
    417
    +0.10
    0.02
    Negative Logits
     done
    -2.01
    alus
    -1.70
     trouble
    -1.63
    hav
    -1.50
     drastic
    -1.49
    lessly
    -1.41
     fundamentally
    -1.39
    ensory
    -1.35
     lax
    -1.33
     hers
    -1.33
    POSITIVE LOGITS
    ħ
    2.25
    ¿
    2.24
    ĸ
    2.16
    µ
    2.13
    »¿
    2.09
    isations
    2.08
    wide
    2.07
    ities
    2.06
    ĵ
    2.06
    ties
    2.02
    Activations Density 0.069%

    No Known Activations