© 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. 1-RES-SM
    11. 26303
    Prev
    Next
    INDEX
    Explanations

    phrases or references related to initiating actions or events

    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
    59
    +0.14
    0.8%
    164
    +0.12
    0.7%
    161
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    164
    +0.14
    0.02
    161
    +0.12
    0.02
    59
    +0.12
    0.02
    Negative Logits
    uries
    -1.82
    ance
    -1.65
    ]--
    -1.59
     Commons
    -1.50
    ivid
    -1.48
    ences
    -1.47
    ...]
    -1.45
    ère
    -1.42
    CoA
    -1.41
    ories
    -1.39
    POSITIVE LOGITS
    starter
    2.87
    off
    2.00
    box
    1.82
    tip
    1.72
     bombs
    1.65
    ified
    1.64
    ref
    1.58
    ged
    1.57
    guns
    1.55
    nel
    1.54
    Activations Density 0.132%

    No Known Activations