© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 4-RES-SM
    11. 9139
    Prev
    Next
    INDEX
    Explanations

    phrases related to offerings and services provided

    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/4-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.4.hook_resid_post
    Hook Layer
    4
    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
    307
    +0.20
    1.1%
    306
    +0.15
    0.9%
    292
    +0.14
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    112
    +0.20
    0.04
    173
    +0.15
    0.04
    292
    +0.14
    0.05
    Negative Logits
    «
    -4.47
    ĨĴ
    -4.28
    İ
    -4.22
    Ļª
    -4.09
    ±
    -4.07
    ¥
    -4.04
    -3.98
    <|outofrange|>
    -3.98
                ↵
    -3.98
    ↵
    -3.98
    POSITIVE LOGITS
     discounts
    2.65
     solutions
    2.28
     alternatives
    2.26
     additional
    2.05
     options
    2.04
     incentives
    2.00
     assistance
    1.98
     alternative
    1.98
     plenty
    1.96
     discounted
    1.95
    Activations Density 0.532%

    No Known Activations