© 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. 12442
    Prev
    Next
    INDEX
    Explanations

    systematic phrases related to conditions or states of an object or entity

    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
    419
    +0.19
    1.1%
    271
    +0.18
    1.1%
    17
    +0.15
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    17
    +0.19
    -0.02
    45
    +0.18
    0.07
    90
    +0.15
    0.11
    Negative Logits
    lei
    -1.76
    askell
    -1.65
    ↵
    -1.42
    plan
    -1.39
    .[@
    -1.39
    taient
    -1.38
    (`
    -1.37
    reens
    -1.36
    .^\[[@
    -1.35
    sta
    -1.34
    POSITIVE LOGITS
    Ĥ¬
    3.23
    ·¸
    3.23
    ĨĴ
    3.22
    ¹
    2.94
    ±
    2.91
    ¬
    2.89
    Į
    2.88
    ¿
    2.82
    Ĩ
    2.75
    §
    2.75
    Activations Density 6.048%

    No Known Activations