© 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 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. 10053
    Prev
    Next
    INDEX
    Explanations

    semicolon characters in code snippets

    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
    478
    +0.21
    1.2%
    350
    +0.13
    0.7%
    8
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    8
    +0.21
    0.05
    86
    +0.13
    0.04
    202
    +0.13
    0.04
    Negative Logits
    ĩ
    -3.45
    ª
    -3.26
    Ī
    -3.20
    ¿½
    -3.19
    ĻĤ
    -3.19
    ģ
    -3.11
    ł
    -3.11
    ®
    -3.10
    ĥ
    -3.06
    ĸ
    -3.01
    POSITIVE LOGITS
     charset
    2.14
    -)
    1.66
    ^](#
    1.53
    def
    1.45
    comes
    1.44
    \!\
    1.43
     filename
    1.39
    charset
    1.39
     file
    1.38
    WITH
    1.36
    Activations Density 0.253%

    No Known Activations