© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 29899
    Prev
    Next
    INDEX
    Explanations

    the letter 'o' in various contexts

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Neuron Alignment
    Index
    Value
    % of L₁
    156
    +0.37
    2.2%
    111
    +0.15
    0.9%
    271
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    111
    +0.37
    0.05
    156
    +0.15
    0.03
    271
    +0.10
    0.01
    Negative Logits
    ories
    -1.54
    oire
    -1.45
    érie
    -1.40
    ppat
    -1.39
     Hide
    -1.35
    ère
    -1.29
     Publ
    -1.25
    ↵
    -1.23
     AFP
    -1.23
    oors
    -1.23
    POSITIVE LOGITS
    ĻĤ
    2.80
    Ļ
    2.75
    »¿
    2.56
    ®
    2.45
    ·¸
    2.39
    ¾
    2.31
    ½
    2.19
    ľ
    2.18
    ¿
    2.13
    ģ
    2.11
    Activations Density 0.061%

    No Known Activations