© 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. 13239
    Prev
    Next
    INDEX
    Explanations

    abbreviations and acronyms commonly used in business or finance contexts

    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
    348
    +0.20
    1.1%
    156
    +0.16
    0.9%
    404
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    369
    +0.20
    0.03
    313
    +0.16
    0.04
    134
    +0.12
    0.07
    Negative Logits
    fulness
    -2.22
    uous
    -2.18
    hood
    -2.11
    ifiable
    -2.02
    aneous
    -2.00
    terday
    -1.99
    ification
    -1.89
    ifice
    -1.83
    ful
    -1.80
    ified
    -1.78
    POSITIVE LOGITS
    Ps
    2.60
    BC
    2.56
    Gs
    2.47
    SE
    2.45
    BM
    2.44
    MC
    2.38
    CC
    2.35
    MS
    2.33
    MT
    2.33
    MA
    2.30
    Activations Density 0.681%

    No Known Activations