© 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. 2-RES-SM
    11. 31422
    Prev
    Next
    INDEX
    Explanations

    terms related to positions, locations, or rankings

    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
    98
    +0.16
    0.9%
    207
    +0.12
    0.7%
    330
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    145
    +0.16
    0.12
    330
    +0.12
    0.07
    155
    +0.11
    0.09
    Negative Logits
     authors
    -1.58
     flavors
    -1.54
     courage
    -1.50
    ľĵ
    -1.49
     citizenship
    -1.47
     things
    -1.45
     Appellant
    -1.44
     Appellants
    -1.44
     chemo
    -1.43
     arrival
    -1.42
    POSITIVE LOGITS
    osphere
    1.96
    wikipedia
    1.85
    itals
    1.81
     vicinity
    1.78
     arsenal
    1.78
    reon
    1.72
     itself
    1.63
     realm
    1.58
     area
    1.58
    pective
    1.55
    Activations Density 0.865%

    No Known Activations