© 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. 30659
    Prev
    Next
    INDEX
    Explanations

    possessive pronouns and terms related to ownership

    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
    156
    +0.16
    0.9%
    240
    +0.13
    0.8%
    338
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.16
    0.08
    405
    +0.13
    0.07
    454
    +0.12
    0.05
    Negative Logits
    ption
    -1.66
    sible
    -1.53
     zones
    -1.47
     itself
    -1.46
     zone
    -1.40
     slot
    -1.35
     capitalism
    -1.34
     arena
    -1.34
     reins
    -1.32
    ament
    -1.32
    POSITIVE LOGITS
    ¨
    3.12
    ĻĤ
    3.02
    ¶
    2.99
    ĺ
    2.97
    Ĺ
    2.95
    ķ
    2.92
    čč
    2.90
    2.90
    ↵
    2.90
    ↵
    2.90
    Activations Density 0.854%

    No Known Activations