© 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. 13709
    Prev
    Next
    INDEX
    Explanations

    technical terms and references related to data management

    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
    156
    +0.26
    1.5%
    17
    +0.15
    0.9%
    88
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    17
    +0.26
    0.00
    88
    +0.15
    0.04
    245
    +0.13
    0.04
    Negative Logits
    ]{}]{}
    -1.44
     warranty
    -1.42
    methods
    -1.40
    ãĥ¼ãĤ¸
    -1.38
    Methods
    -1.36
    )\].
    -1.36
    PHY
    -1.35
    容
    -1.31
    BSD
    -1.29
    alities
    -1.28
    POSITIVE LOGITS
     doubles
    1.50
    eries
    1.48
    chers
    1.45
     tennis
    1.42
     CSV
    1.41
     bras
    1.40
     lottery
    1.40
     WHERE
    1.38
     XY
    1.35
    ples
    1.33
    Activations Density 0.535%

    No Known Activations