© 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. 1-RES-SM
    11. 29130
    Prev
    Next
    INDEX
    Explanations

    terms and attributes related to network structures and data management

    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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    156
    +0.29
    1.7%
    143
    +0.12
    0.7%
    274
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    143
    +0.29
    0.14
    17
    +0.12
    0.07
    379
    +0.11
    0.15
    Negative Logits
    ´
    -2.50
    Ļ
    -2.38
    «
    -2.19
    ·
    -2.14
    Ĩ
    -2.10
    ħ
    -2.07
    ĺ
    -2.00
    ¿½
    -1.98
    -1.96
    ↵  ↵
    -1.96
    POSITIVE LOGITS
     Unless
    1.64
    fonts
    1.63
     Without
    1.53
     Redist
    1.47
     Tonight
    1.42
    ":
    1.40
    nier
    1.40
     Luck
    1.37
    walk
    1.36
     Whenever
    1.35
    Activations Density 3.100%

    No Known Activations