© 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. Residuals
    5. 2-RES-SM
    6. 29892
    Prev
    Next
    INDEX
    Explanations

    expressions of uncertainty or speculation

    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
    478
    +0.15
    0.8%
    442
    +0.11
    0.6%
    94
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    94
    +0.15
    0.03
    219
    +0.11
    0.03
    33
    +0.11
    0.04
    Negative Logits
    liness
    -1.68
    ership
    -1.67
    AP
    -1.59
     process
    -1.52
    HS
    -1.51
    ÃŁ
    -1.51
    ÃŁe
    -1.51
     Himself
    -1.50
    ocks
    -1.47
     alike
    -1.45
    POSITIVE LOGITS
     someday
    1.86
    olia
    1.66
    GREEK
    1.65
     luck
    1.59
     elif
    1.55
     Ctrl
    1.46
     "!"
    1.43
     another
    1.40
     @@
    1.40
     joking
    1.39
    Activations Density 0.190%

    No Known Activations