© 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
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. Residuals
    5. 4-RES-SM
    6. 12116
    Prev
    Next
    INDEX
    Explanations

    mentions of page counts or references to pagination

    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
    156
    +0.48
    3.1%
    111
    +0.28
    1.8%
    510
    +0.12
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    419
    +0.48
    0.07
    40
    +0.28
    0.08
    76
    +0.12
    0.08
    Negative Logits
    ĥ½
    -6.50
    <|outofrange|>
    -6.41
    ↵↵                               
    -6.41
    ↵↵            
    -6.41
     ↵    
    -6.41
                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                    
    -6.41
    ↵  ³³³
    -6.41
    ↵                                                
    -6.41
           ↵ 
    -6.41
    <|outofrange|>
    -6.41
    POSITIVE LOGITS
    wers
    1.57
    pread
    1.48
    quare
    1.46
    eded
    1.42
    es
    1.40
    pace
    1.39
    elen
    1.38
    eted
    1.34
    blogspot
    1.33
    ety
    1.32
    Activations Density 0.120%

    No Known Activations