© 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
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. Residuals
    5. 2-RES-SM
    6. 28108
    Prev
    Next
    INDEX
    Explanations

    programming keywords and syntax related to exceptions and method definitions

    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
    76
    +0.18
    1.0%
    218
    +0.16
    0.9%
    117
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    465
    +0.18
    0.17
    407
    +0.16
    0.11
    282
    +0.13
    0.11
    Negative Logits
    ENCE
    -1.44
    itud
    -1.44
     read
    -1.36
    ITION
    -1.33
     preced
    -1.30
     absorbance
    -1.30
    IRE
    -1.30
     pertains
    -1.29
    culation
    -1.29
    ITLE
    -1.29
    POSITIVE LOGITS
    Ļª
    1.62
    »¿
    1.56
    izon
    1.55
    ober
    1.46
    googleapis
    1.41
    ollo
    1.41
    gol
    1.39
    ço
    1.37
    zez
    1.35
    kot
    1.33
    Activations Density 4.834%

    No Known Activations