© 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. 30828
    Prev
    Next
    INDEX
    Explanations

    legal terms associated with software licenses

    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
    145
    +0.13
    0.7%
    186
    +0.11
    0.7%
    414
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    145
    +0.13
    0.02
    7
    +0.11
    0.01
    74
    +0.10
    0.01
    Negative Logits
     lining
    -1.51
    blast
    -1.50
    '):
    -1.45
    ect
    -1.40
    apse
    -1.38
    separator
    -1.36
    )!
    -1.32
    ectomy
    -1.31
    oscopy
    -1.31
    uminescence
    -1.29
    POSITIVE LOGITS
    rant
    1.52
     WARRANTIES
    1.41
    smen
    1.37
    Ùİ
    1.30
    ÙĬÙĨ
    1.25
    اÙĨ
    1.25
    pson
    1.23
    gments
    1.22
     extrinsic
    1.21
    rants
    1.21
    Activations Density 0.075%

    No Known Activations