© 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. 32091
    Prev
    Next
    INDEX
    Explanations

    URLs and web links

    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
    261
    +0.12
    0.7%
    289
    +0.11
    0.6%
    104
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    82
    +0.12
    0.03
    315
    +0.11
    0.03
    261
    +0.11
    0.03
    Negative Logits
     alike
    -1.59
     themselves
    -1.52
    isation
    -1.52
    thems
    -1.50
     defence
    -1.40
     genomes
    -1.38
     succession
    -1.37
     (§
    -1.37
     itself
    -1.37
     worldwide
    -1.37
    POSITIVE LOGITS
    "}](#
    1.54
    ä¹Ī
    1.46
    ZZ
    1.37
    RY
    1.34
    dz
    1.34
    ento
    1.34
    >\
    1.33
    010
    1.30
    enz
    1.30
    LLOW
    1.29
    Activations Density 0.428%

    No Known Activations