© 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. 2-RES-SM
    11. 12782
    Prev
    Next
    INDEX
    Explanations

    references to the Mozilla Public License and its terms

    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
    222
    +0.12
    0.7%
    499
    +0.12
    0.7%
    102
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    222
    +0.12
    0.03
    329
    +0.12
    0.02
    103
    +0.11
    0.03
    Negative Logits
    ®
    -4.42
    ĨĴ
    -4.36
    ¦
    -4.22
    §
    -4.15
    ³
    -4.05
    ļ
    -4.03
    ĵ
    -4.02
    ¯
    -4.02
    ł
    -4.00
    ¢
    -3.99
    POSITIVE LOGITS
     COPYRIGHT
    1.71
     license
    1.65
    atype
    1.63
     Redistributions
    1.63
    uto
    1.62
     cite
    1.58
     permission
    1.57
    urname
    1.51
    unto
    1.50
    NdEx
    1.50
    Activations Density 0.180%

    No Known Activations