© 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. 24812
    Prev
    Next
    INDEX
    Explanations

    occurrences of the character '/' in the document

    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
    265
    +0.14
    0.8%
    114
    +0.13
    0.8%
    481
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    114
    +0.14
    0.03
    265
    +0.13
    0.02
    214
    +0.11
    0.01
    Negative Logits
    ...](
    -1.52
     elapsed
    -1.45
     favorably
    -1.43
     tuned
    -1.43
     wisely
    -1.42
     differently
    -1.35
     seam
    -1.34
     correctly
    -1.33
    ocyte
    -1.30
    hing
    -1.30
    POSITIVE LOGITS
    amily
    1.77
    microsoft
    1.66
    docs
    1.64
    mozilla
    1.47
    facebook
    1.44
    atories
    1.44
    ennes
    1.41
    intern
    1.38
    npmjs
    1.38
    ductory
    1.37
    Activations Density 0.076%

    No Known Activations