© 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. 29231
    Prev
    Next
    INDEX
    Explanations

    sentences ending with a punctuation mark

    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
    258
    +0.18
    1.1%
    203
    +0.18
    1.1%
    369
    +0.16
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    258
    +0.18
    0.12
    203
    +0.18
    0.12
    111
    +0.16
    0.13
    Negative Logits
    noreply
    -1.65
    ermann
    -1.51
    blr
    -1.49
    asures
    -1.44
    erton
    -1.42
    EventListener
    -1.40
    à±ģ
    -1.40
    osomes
    -1.39
     chat
    -1.37
    ular
    -1.36
    POSITIVE LOGITS
    Ĥ
    2.05
    ©
    2.04
    »
    2.00
    §
    1.93
    Ĵ
    1.89
    ½
    1.82
    ī
    1.80
    ģ
    1.79
    ³
    1.78
    º
    1.75
    Activations Density 1.787%

    No Known Activations