© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 28042
    Prev
    Next
    INDEX
    Explanations

    terms indicating variation or difference

    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
    457
    +0.14
    0.8%
    231
    +0.12
    0.6%
    156
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    457
    +0.14
    0.04
    231
    +0.12
    0.03
    29
    +0.11
    0.02
    Negative Logits
    =====
    -1.67
    ISE
    -1.65
    ````
    -1.51
    bers
    -1.49
    =========
    -1.49
    ---|---|---
    -1.44
    eks
    -1.42
    fter
    -1.42
    igation
    -1.42
    ========
    -1.42
    POSITIVE LOGITS
     kinds
    2.29
     sized
    2.26
    iates
    2.25
    iator
    2.18
    iable
    2.12
    ?).
    2.04
     types
    2.03
     sizes
    1.99
    ials
    1.99
     sorts
    1.97
    Activations Density 0.108%

    No Known Activations