© 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
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 2-RES-SM
    11. 12311
    Prev
    Next
    INDEX
    Explanations

    phrases that indicate uncertainty or conditionality in statements

    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
    36
    +0.12
    0.6%
    149
    +0.11
    0.6%
    33
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    149
    +0.12
    0.07
    88
    +0.11
    0.07
    496
    +0.11
    0.07
    Negative Logits
     yourself
    -1.85
     air
    -1.70
     lunch
    -1.60
     myself
    -1.57
     breeze
    -1.53
     ride
    -1.50
     Aviation
    -1.47
     me
    -1.47
     Elementary
    -1.45
     himself
    -1.44
    POSITIVE LOGITS
     ↵
    4.15
     ↵
    4.15
    4.15
           ↵
    4.15
     ↵
    4.15
    ↵
    4.15
    č↵
    4.15
    <|outofrange|>
    4.15
    č↵
    4.15
    ↵
    4.15
    Activations Density 0.462%

    No Known Activations