© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 4-RES-SM
    11. 12654
    Prev
    Next
    INDEX
    Explanations

    references to programming languages, frameworks, and libraries

    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/4-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.4.hook_resid_post
    Hook Layer
    4
    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
    78
    +0.19
    1.2%
    348
    +0.18
    1.1%
    466
    +0.16
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    78
    +0.19
    -0.00
    334
    +0.18
    0.05
    59
    +0.16
    0.05
    Negative Logits
    ergic
    -1.62
    ationally
    -1.51
    aterally
    -1.44
    above
    -1.42
     herein
    -1.37
     composition
    -1.37
     compositions
    -1.32
     hereto
    -1.31
     contained
    -1.26
    coma
    -1.26
    POSITIVE LOGITS
    Ń
    2.36
    ¤
    2.27
    į
    2.13
    º
    2.12
    «
    2.01
    ¡
    1.98
    ¾
    1.96
    §
    1.96
    ĸ
    1.94
    ¿
    1.94
    Activations Density 0.403%

    No Known Activations