© 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. 28728
    Prev
    Next
    INDEX
    Explanations

    references to licenses and their terms

    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
    478
    +0.14
    0.9%
    167
    +0.11
    0.7%
    320
    +0.11
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    116
    +0.14
    0.03
    167
    +0.11
    0.02
    464
    +0.11
    0.02
    Negative Logits
    ²
    -1.59
     scene
    -1.42
     crore
    -1.36
    ł
    -1.29
    ember
    -1.29
    Ł
    -1.28
    ¶
    -1.27
    ¦
    -1.26
    »
    -1.25
    Errno
    -1.25
    POSITIVE LOGITS
    iday
    1.51
    iah
    1.49
     Redistributions
    1.48
     deference
    1.43
    ableness
    1.39
    arten
    1.34
    ween
    1.31
    ifts
    1.25
     nonetheless
    1.25
     generally
    1.24
    Activations Density 0.118%

    No Known Activations