© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 1-RES-SM
    11. 23796
    Prev
    Next
    INDEX
    Explanations

    references to states or conditions

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__res-sm_processed/1-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.1.hook_resid_post
    Hook Layer
    1
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Neuron Alignment
    Index
    Value
    % of L₁
    199
    +0.13
    0.7%
    302
    +0.11
    0.6%
    503
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    199
    +0.13
    0.04
    503
    +0.11
    0.02
    457
    +0.11
    0.03
    Negative Logits
     Redistributions
    -1.50
    estones
    -1.49
     strides
    -1.45
    fully
    -1.43
    raiser
    -1.40
    lessly
    -1.35
     reliance
    -1.34
    hops
    -1.33
     RIP
    -1.29
    length
    -1.28
    POSITIVE LOGITS
    ĵ
    3.15
    Ł
    3.07
    ¸
    2.95
    ķ
    2.92
    ¿½
    2.85
    ij
    2.85
    £
    2.75
    ©
    2.71
    Ĵ
    2.67
    Ń
    2.67
    Activations Density 0.498%

    No Known Activations