© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 4-RES-SM
    11. 11882
    Prev
    Next
    INDEX
    Explanations

    numeric values, particularly in a format that resembles data or identifiers

    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/4-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.4.hook_resid_post
    Hook Layer
    4
    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
    +0.39
    2.3%
    136
    +0.20
    1.2%
    271
    +0.20
    1.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    333
    +0.39
    0.11
    136
    +0.20
    -0.03
    502
    +0.20
    0.11
    Negative Logits
    matically
    -1.61
    ially
    -1.49
    men
    -1.26
     fault
    -1.24
    resses
    -1.23
    ress
    -1.23
     Body
    -1.22
     Anonymous
    -1.22
     mine
    -1.20
     practitioner
    -1.20
    POSITIVE LOGITS
    ĨĴ
    2.31
    «
    2.21
    »¿
    2.17
    ©
    2.16
    "}](#
    2.08
    ·¸
    2.06
    ĻĤ
    2.05
    ^](#
    2.03
    )}=\
    2.01
    ĭ
    1.98
    Activations Density 1.734%

    No Known Activations