© 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. 12831
    Prev
    Next
    INDEX
    Explanations

    technical terms and coding references, specifically related to programming

    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
    31
    +0.15
    0.9%
    320
    +0.14
    0.8%
    127
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    31
    +0.15
    0.18
    178
    +0.14
    0.13
    278
    +0.13
    0.06
    Negative Logits
    ↵
    -1.68
    nae
    -1.52
    zzi
    -1.47
    ños
    -1.44
    Äĩ
    -1.37
     Weinstein
    -1.34
    tees
    -1.28
    nikov
    -1.26
    ↵↵
    -1.26
    hua
    -1.23
    POSITIVE LOGITS
     oath
    1.61
    ~\
    1.58
     format
    1.52
    enic
    1.44
    CDATA
    1.36
    ameter
    1.31
    UNIT
    1.29
    ophosph
    1.25
    @
    1.25
    ceptive
    1.24
    Activations Density 5.671%

    No Known Activations