© 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. 1-RES-SM
    11. 25377
    Prev
    Next
    INDEX
    Explanations

    references to locations or places within the text

    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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    325
    +0.12
    0.7%
    132
    +0.12
    0.6%
    201
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    198
    +0.12
    0.03
    132
    +0.12
    0.04
    325
    +0.11
    0.03
    Negative Logits
    »¿
    -1.79
    ĺ
    -1.79
    Ī
    -1.78
    Ħ
    -1.76
    Į
    -1.74
    §
    -1.73
    ¶
    -1.72
    Ŀ
    -1.72
    ¥
    -1.65
    «
    -1.64
    POSITIVE LOGITS
    holder
    2.21
    bos
    1.97
    unov
    1.80
    away
    1.73
    holders
    1.70
    finder
    1.67
     mixer
    1.62
    aria
    1.60
    amps
    1.60
    van
    1.58
    Activations Density 0.106%

    No Known Activations