© 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. 1-RES-SM
    6. 26687
    Prev
    Next
    INDEX
    Explanations

    instances of the word "contrast."

    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
    335
    +0.12
    0.7%
    248
    +0.10
    0.6%
    380
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    367
    +0.12
    0.01
    335
    +0.10
    0.01
    485
    +0.10
    0.01
    Negative Logits
    ½
    -4.32
    IJ
    -4.20
    ĻĤ
    -4.12
    ±
    -4.10
    »¿
    -4.08
    Ļ
    -4.03
    ¾
    -3.98
    Ĥ
    -3.97
    Ħ
    -3.89
    ¿½
    -3.86
    POSITIVE LOGITS
    thouse
    2.09
    iveness
    2.03
    ively
    1.89
    rup
    1.75
    ivity
    1.74
    ibility
    1.70
    inho
    1.68
     screening
    1.65
    acting
    1.64
    yp
    1.63
    Activations Density 0.062%

    No Known Activations