© 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. 10354
    Prev
    Next
    INDEX
    Explanations

    the name "Sam"

    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
    +0.20
    1.1%
    369
    +0.17
    1.0%
    213
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    410
    +0.20
    0.03
    369
    +0.17
    0.01
    177
    +0.13
    0.02
    Negative Logits
    americ
    -1.81
     behalf
    -1.64
     course
    -1.59
    )|$(
    -1.53
    FORE
    -1.46
    office
    -1.45
    indexOf
    -1.44
    closing
    -1.44
    iquit
    -1.43
    course
    -1.42
    POSITIVE LOGITS
    urai
    2.67
    antha
    2.44
    pler
    2.26
    pson
    2.24
    uel
    1.90
    iewicz
    1.90
    uli
    1.74
    owitz
    1.71
    ettes
    1.65
    iti
    1.61
    Activations Density 0.148%

    No Known Activations