Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. Residuals
    5. 2-RES-SM
    6. 29488
    Prev
    Next
    INDEX
    Explanations

    legal references and case details

    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
    265
    +0.12
    0.7%
    145
    +0.11
    0.6%
    447
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    319
    +0.12
    0.03
    34
    +0.11
    0.04
    145
    +0.10
    0.04
    Negative Logits
     respectively
    -1.89
    iona
    -1.54
    oche
    -1.54
    ,...
    -1.50
     etc
    -1.50
     doesnt
    -1.46
    ctomy
    -1.46
    cdot
    -1.45
    ?!
    -1.45
    â̦
    -1.43
    POSITIVE LOGITS
    ĥ½
    3.07
          ↵ 
    2.87
                                     
    2.87
                                                          
    2.87
    <|outofrange|>
    2.87
    ↵                                 
    2.87
                                                     
    2.87
    ↵     
    2.87
    ↵  ³³³
    2.87
    ↵                                                                                                
    2.87
    Activations Density 0.252%

    No Known Activations