© 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. 12741
    Prev
    Next
    INDEX
    Explanations

    phrases related to conditions and qualifications that affect outcomes

    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
    499
    +0.12
    0.7%
    465
    +0.11
    0.6%
    424
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    176
    +0.12
    0.08
    95
    +0.11
    0.07
    492
    +0.10
    0.06
    Negative Logits
    ŀ
    -2.00
    ¯
    -1.96
    ±
    -1.86
    Ļ
    -1.80
    ¿½
    -1.76
    ¢
    -1.75
    ²
    -1.73
    º
    -1.69
    ³
    -1.62
    °
    -1.59
    POSITIVE LOGITS
     direction
    2.04
     circumstance
    1.72
     circumstances
    1.68
    ihad
    1.57
     dtype
    1.53
    yll
    1.52
     keywords
    1.51
    edia
    1.48
    ço
    1.47
     basis
    1.46
    Activations Density 1.281%

    No Known Activations