© 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. 29727
    Prev
    Next
    INDEX
    Explanations

    references to locations and affiliations

    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
    156
    +0.32
    1.8%
    488
    +0.12
    0.7%
    47
    +0.11
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    47
    +0.32
    0.10
    90
    +0.12
    0.09
    156
    +0.11
    0.11
    Negative Logits
    -2.51
    -2.51
    -2.51
    ↵            ↵
    -2.51
           ↵
    -2.51
     ↵
    -2.51
    ↵↵
    -2.51
    <|outofrange|>
    -2.51
    ↵
    -2.51
    č↵
    -2.51
    POSITIVE LOGITS
    ivat
    1.63
    imar
    1.53
    etine
    1.51
    vas
    1.48
    rosse
    1.45
     Angeles
    1.44
    isl
    1.41
    ony
    1.41
    amba
    1.39
    ynaptic
    1.39
    Activations Density 1.743%

    No Known Activations