© 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. 4-RES-SM
    11. 8578
    Prev
    Next
    INDEX
    Explanations

    details about property features and configurations

    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/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
    Neuron Alignment
    Index
    Value
    % of L₁
    412
    +0.19
    1.1%
    199
    +0.12
    0.7%
    292
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    412
    +0.19
    0.15
    368
    +0.12
    0.05
    195
    +0.12
    0.10
    Negative Logits
    «
    -2.64
    ·¸
    -2.63
    į
    -2.61
    ł
    -2.56
    ¨
    -2.54
    Ķ
    -2.48
    -2.47
    -2.47
    č↵č↵
    -2.47
    -2.47
    POSITIVE LOGITS
     rooms
    3.00
     amenities
    2.97
     bedrooms
    2.86
     Hotel
    2.71
     upstairs
    2.65
     accommodation
    2.59
     guests
    2.59
     apartments
    2.54
     hotel
    2.54
     bedroom
    2.53
    Activations Density 1.734%

    No Known Activations