© 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
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. Residuals
    5. 4-RES-SM
    6. 11463
    Prev
    Next
    INDEX
    Explanations

    concepts related to mathematical and structural properties, particularly in the context of consistency and reliability in various systems

    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/4-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.4.hook_resid_post
    Hook Layer
    4
    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
    232
    +0.13
    0.7%
    198
    +0.12
    0.7%
    14
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    482
    +0.13
    0.04
    198
    +0.12
    0.07
    319
    +0.11
    0.05
    Negative Logits
    º
    -5.22
    Ĥ¬
    -5.13
    ±
    -5.11
    Ĥ
    -5.09
    ¼
    -5.09
           ↵ 
    -4.97
    ↵	  
    -4.97
                                                                    
    -4.97
              ↵ 
    -4.97
                                                      
    -4.97
    POSITIVE LOGITS
     criterion
    2.30
     traits
    2.20
     criteria
    2.14
     requirements
    2.12
     rating
    2.08
     reasons
    2.07
     issues
    2.06
     ratings
    2.02
     constraints
    1.99
     constraint
    1.98
    Activations Density 0.674%

    No Known Activations