© 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. 27437
    Prev
    Next
    INDEX
    Explanations

    punctuation marks, particularly quotes and apostrophes

    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
    +0.54
    3.8%
    111
    +0.39
    2.7%
    271
    +0.10
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    156
    +0.54
    0.41
    71
    +0.39
    0.38
    111
    +0.10
    0.40
    Negative Logits
    stown
    -1.56
    ¾
    -1.40
    pired
    -1.32
    ķ
    -1.28
     ensued
    -1.18
    ...](
    -1.17
    <|outofrange|>
    -1.16
    -1.16
    ↵
    -1.16
    ↵↵
    -1.16
    POSITIVE LOGITS
    msgstr
    1.39
     recess
    1.35
    =>
    1.30
    emitting
    1.29
     =>
    1.28
    être
    1.23
     msgid
    1.17
    rbrace
    1.17
    Undecided
    1.17
    ubot
    1.16
    Activations Density 1.230%

    No Known Activations