© 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. 28328
    Prev
    Next
    INDEX
    Explanations

    terms related to digital or technical discussions

    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
    12
    +0.15
    0.9%
    417
    +0.13
    0.8%
    186
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    417
    +0.15
    0.13
    221
    +0.13
    0.08
    319
    +0.13
    0.07
    Negative Logits
     );
    -1.71
    ypse
    -1.67
     hers
    -1.57
     respectively
    -1.56
      ↵
    -1.51
     okay
    -1.50
     ))
    -1.42
     Redistributions
    -1.40
    ")]
    -1.38
    estamp
    -1.37
    POSITIVE LOGITS
    ł
    3.60
    ¹
    3.59
    ³
    3.56
    Ł
    3.52
    ²
    3.52
    ¼
    3.38
    ½
    3.35
    ¸
    3.29
    ©
    3.25
    ļ
    3.24
    Activations Density 4.079%

    No Known Activations