© 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. 28373
    Prev
    Next
    INDEX
    Explanations

    specific numeric citations or references in legal or formal contexts

    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
    261
    +0.18
    1.1%
    263
    +0.14
    0.8%
    34
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    116
    +0.18
    0.09
    261
    +0.14
    0.05
    83
    +0.13
    0.06
    Negative Logits
    ĥ½
    -3.31
    µ
    -2.56
    ľĵ
    -2.46
    Ŀ
    -2.33
    Ī
    -2.33
    ¸
    -2.22
    £
    -2.12
    ĭ
    -2.09
    ¤
    -2.04
    ¥
    -2.02
    POSITIVE LOGITS
    itely
    1.65
     them
    1.50
     Justice
    1.42
     EF
    1.40
     thread
    1.39
     ART
    1.39
    -->
    1.38
    ://
    1.36
    them
    1.33
     HA
    1.29
    Activations Density 4.579%

    No Known Activations