© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 12143
    Prev
    Next
    INDEX
    Explanations

    references to divisions or categorizations within a group or system

    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
    307
    +0.15
    0.9%
    466
    +0.12
    0.6%
    447
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    364
    +0.15
    0.03
    296
    +0.12
    0.03
    98
    +0.12
    0.01
    Negative Logits
    ®
    -2.55
    Ĩ
    -2.31
    ĺ
    -2.11
    »
    -2.10
    ²
    -2.09
    »¿
    -2.02
    ½
    -2.01
    «
    -2.01
    Ł
    -1.96
    Ĥ
    -1.96
    POSITIVE LOGITS
     categories
    2.63
     groups
    2.51
     subgroups
    2.50
     two
    2.36
     three
    2.23
     four
    2.22
     segments
    2.15
     Groups
    2.08
     ranks
    2.01
     tert
    2.00
    Activations Density 0.340%

    No Known Activations