© 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. 12007
    Prev
    Next
    INDEX
    Explanations

    references to status codes and operational states

    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
    9
    +0.17
    1.0%
    172
    +0.15
    0.9%
    397
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    9
    +0.17
    0.02
    397
    +0.15
    0.02
    233
    +0.14
    0.02
    Negative Logits
    keit
    -1.95
    inary
    -1.67
    utz
    -1.61
    nai
    -1.60
     LLP
    -1.58
    amous
    -1.52
    ry
    -1.49
    plify
    -1.49
    cted
    -1.46
     diameter
    -1.46
    POSITIVE LOGITS
     quo
    4.37
    es
    3.13
     epile
    3.09
    ħ
    2.32
    quo
    2.30
    Ļ
    2.27
    ful
    2.07
    ¦
    2.05
    ·
    2.03
    guard
    2.01
    Activations Density 0.342%

    No Known Activations