© 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. 2-RES-SM
    11. 9879
    Prev
    Next
    INDEX
    Explanations

    XML namespace definitions

    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/2-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.2.hook_resid_post
    Hook Layer
    2
    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
    479
    +0.12
    0.7%
    464
    +0.12
    0.7%
    71
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    74
    +0.12
    0.01
    18
    +0.12
    0.01
    375
    +0.12
    0.01
    Negative Logits
     lies
    -1.51
    mort
    -1.49
    zia
    -1.48
     deaths
    -1.41
     lately
    -1.39
    ons
    -1.35
     average
    -1.34
    ship
    -1.34
    abouts
    -1.33
     misplaced
    -1.32
    POSITIVE LOGITS
    ©
    1.58
    ¦
    1.52
    ³
    1.49
    ="../../../../../../
    1.47
    ¤
    1.46
    ¸
    1.44
    Īĺ
    1.42
    Ĥ¬
    1.38
    ellation
    1.38
    ermal
    1.37
    Activations Density 0.074%

    No Known Activations