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. MLP Post
    5. 0-MLP-SM
    6. 27247
    Prev
    Next
    INDEX
    Explanations

    repeated phrases or sequences that contain the word "ERY"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__mlp-sm_processed/0-mlp-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.0.hook_mlp_out
    Hook Layer
    0
    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
    300
    +0.10
    0.6%
    18
    +0.10
    0.6%
    389
    +0.09
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    300
    +0.10
    0.01
    320
    +0.10
    0.01
    78
    +0.09
    0.01
    Negative Logits
     respectively
    -1.70
    uminous
    -1.70
     existence
    -1.58
     studied
    -1.51
    omitempty
    -1.46
     moderate
    -1.45
    itter
    -1.44
    respectively
    -1.41
    ages
    -1.38
     Minor
    -1.37
    POSITIVE LOGITS
    º
    2.16
    ↵                                 
    2.08
    ↵                                               
    2.08
    ↵                                                                
    2.08
    <|outofrange|>
    2.08
                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                    
    2.08
    <|padding|>
    2.08
    č↵        
    2.08
    ↵  ³³³
    2.08
             
    2.08
    Activations Density 0.014%

    No Known Activations