© 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. 2-RES-SM
    11. 9889
    Prev
    Next
    INDEX
    Explanations

    important terms related to community engagement and action

    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/2-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.2.hook_resid_post
    Hook Layer
    2
    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
    248
    +0.15
    0.9%
    123
    +0.12
    0.7%
    326
    +0.11
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    248
    +0.15
    0.13
    34
    +0.12
    0.15
    95
    +0.11
    0.10
    Negative Logits
     gloves
    -1.68
     socks
    -1.66
     needles
    -1.63
     mice
    -1.59
     hints
    -1.58
     Mice
    -1.58
     lightly
    -1.57
     cords
    -1.56
     arrows
    -1.53
     tubing
    -1.52
    POSITIVE LOGITS
    ductor
    1.64
    oracle
    1.57
     lobby
    1.50
    cephal
    1.47
     electorate
    1.45
    ivist
    1.45
     influential
    1.44
     localities
    1.43
    onen
    1.35
     decision
    1.34
    Activations Density 4.181%

    No Known Activations