© 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. 1-RES-SM
    11. 27077
    Prev
    Next
    INDEX
    Explanations

    specific years that indicate historical events or milestones

    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
    110
    +0.17
    1.0%
    475
    +0.15
    0.8%
    502
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    51
    +0.17
    0.06
    475
    +0.15
    0.07
    308
    +0.13
    0.06
    Negative Logits
    ĻĤ
    -2.60
    ¶
    -2.60
    Ģ
    -2.19
    ·
    -2.17
    ī
    -2.17
    ©
    -2.11
    ¼
    -2.10
              ↵
    -2.07
    č↵
    -2.07
     ↵
    -2.07
    POSITIVE LOGITS
     onwards
    1.74
    ^[@
    1.55
     pronounced
    1.52
     reflected
    1.52
     "...
    1.51
    /'
    1.51
     *"
    1.49
     ([@
    1.49
     renamed
    1.48
     involving
    1.48
    Activations Density 0.184%

    No Known Activations