© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. MLP Post
    5. 2-MLP-SM
    6. 25850
    Prev
    Next
    INDEX
    Explanations

    forms of the word "esophagus."

    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/2-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.2.hook_mlp_out
    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
    156
    +0.24
    1.4%
    23
    +0.15
    0.9%
    483
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    23
    +0.24
    0.03
    374
    +0.15
    0.01
    133
    +0.12
    0.04
    Negative Logits
    ↵                                               
    -2.69
    ↵↵           
    -2.69
    <|outofrange|>
    -2.69
             
    -2.69
    <|outofrange|>
    -2.69
     ↵   
    -2.69
    <|outofrange|>
    -2.69
    <|outofrange|>
    -2.69
    ↵	   
    -2.69
    ↵                                
    -2.69
    POSITIVE LOGITS
    cing
    1.46
    thon
    1.44
     longer
    1.42
     views
    1.41
     proceedings
    1.34
     ($
    1.33
     depths
    1.33
    chair
    1.31
    beth
    1.29
     veins
    1.29
    Activations Density 0.556%

    No Known Activations