© 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. 11543
    Prev
    Next
    INDEX
    Explanations

    past tense verbs and their variations

    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
    156
    +0.21
    1.2%
    348
    +0.15
    0.9%
    493
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    493
    +0.21
    0.07
    162
    +0.15
    0.02
    253
    +0.14
    0.07
    Negative Logits
     needed
    -1.52
    ably
    -1.41
    where
    -1.30
     reflected
    -1.25
     °
    -1.24
     required
    -1.24
    нÑĭм
    -1.24
    usement
    -1.24
    pert
    -1.24
    Did
    -1.23
    POSITIVE LOGITS
    own
    2.74
    ocs
    2.59
    irectory
    2.52
    river
    2.47
    irect
    2.40
    ifferent
    2.35
    icine
    2.27
    etect
    2.25
    irector
    2.23
    ictor
    2.18
    Activations Density 0.626%

    No Known Activations