© 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. 1-RES-SM
    11. 28424
    Prev
    Next
    INDEX
    Explanations

    narratives related to personal experiences or testimonials

    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/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
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    156
    +0.17
    0.9%
    410
    +0.12
    0.7%
    352
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    371
    +0.17
    0.04
    38
    +0.12
    0.03
    8
    +0.11
    0.03
    Negative Logits
    µ
    -2.05
    ĥ½
    -1.96
    ¥
    -1.93
    ¶
    -1.92
    §
    -1.87
    ¯
    -1.83
    ī
    -1.78
    ↵  ↵
    -1.78
    <|outofrange|>
    -1.78
    ↵↵
    -1.78
    POSITIVE LOGITS
    telling
    2.33
    lines
    2.15
    pin
    2.07
    book
    2.05
    boards
    1.98
    piece
    1.92
    pins
    1.91
    line
    1.90
    grounds
    1.87
     behind
    1.79
    Activations Density 0.248%

    No Known Activations