© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. Residuals
    5. 2-RES-SM
    6. 10866
    Prev
    Next
    INDEX
    Explanations

    references to plastic materials and their types

    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
    156
    +0.22
    1.3%
    19
    +0.12
    0.7%
    2
    +0.09
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    443
    +0.22
    0.01
    19
    +0.12
    0.02
    18
    +0.09
    0.01
    Negative Logits
     yours
    -1.69
     favor
    -1.66
     morn
    -1.61
     hers
    -1.59
     midnight
    -1.56
     distant
    -1.56
     awake
    -1.55
     impatient
    -1.52
    neh
    -1.49
     place
    -1.45
    POSITIVE LOGITS
    ¢
    3.07
    Ĵ
    3.00
    ĸ
    2.95
    ¿½
    2.87
    ł
    2.68
    Ħ
    2.65
    ·¸
    2.52
    ĻĤ
    2.52
    ¬
    2.51
    ĵ
    2.50
    Activations Density 0.099%

    No Known Activations