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. 1-RES-SM
    6. 27060
    Prev
    Next
    INDEX
    Explanations

    various forms of the word "op."

    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
    47
    +0.16
    0.9%
    183
    +0.13
    0.7%
    278
    +0.11
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    47
    +0.16
    0.11
    136
    +0.13
    -0.06
    414
    +0.11
    0.08
    Negative Logits
     sworn
    -1.53
     sure
    -1.43
     remed
    -1.38
    then
    -1.38
     Develop
    -1.38
     locals
    -1.36
     Authors
    -1.35
    fiddle
    -1.35
     County
    -1.34
     Works
    -1.33
    POSITIVE LOGITS
    ¼
    4.14
    ·
    4.05
    ↵                                            
    4.00
                                            
    4.00
    <|outofrange|>
    4.00
    ↵↵    
    4.00
    ↵  âĢĥ
    4.00
    ↵↵                       
    4.00
                              
    4.00
    ↵                                             
    4.00
    Activations Density 2.755%

    No Known Activations