© 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
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 2-RES-SM
    11. 30245
    Prev
    Next
    INDEX
    Explanations

    references to the action of removing or eliminating items

    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
    287
    +0.13
    0.7%
    176
    +0.13
    0.7%
    376
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    287
    +0.13
    0.04
    239
    +0.13
    0.04
    148
    +0.11
    0.01
    Negative Logits
    ĵ
    -2.58
    ĥ
    -2.40
    §
    -2.12
    ķ
    -1.97
    Īĺ
    -1.96
    ¦
    -1.91
    Ĺ
    -1.85
    Ļ
    -1.83
    Ī
    -1.81
    Ĵ
    -1.74
    POSITIVE LOGITS
     duplicates
    1.85
     altogether
    1.81
     impurities
    1.81
     throat
    1.76
     unwanted
    1.61
     tabs
    1.61
    stituted
    1.61
     offending
    1.60
    Rows
    1.57
     sins
    1.55
    Activations Density 0.165%

    No Known Activations