© 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
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. Residuals
    5. 4-RES-SM
    6. 11570
    Prev
    Next
    INDEX
    Explanations

    mentions of offerings, services, or provisions in the context of health and safety

    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/4-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.4.hook_resid_post
    Hook Layer
    4
    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
    307
    +0.27
    1.5%
    466
    +0.13
    0.7%
    472
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    69
    +0.27
    0.11
    485
    +0.13
    0.06
    307
    +0.10
    0.12
    Negative Logits
    ↵							
    -1.54
    not
    -1.41
    rather
    -1.38
    ache
    -1.35
     briefly
    -1.33
     jointly
    -1.27
    ät
    -1.25
    TON
    -1.24
    ®
    -1.24
    still
    -1.23
    POSITIVE LOGITS
     anymore
    4.55
     anything
    4.48
     any
    4.47
     nor
    4.06
     anywhere
    3.43
     anyone
    3.38
     anybody
    3.24
     slightest
    3.13
     whatsoever
    3.05
     ANY
    3.05
    Activations Density 1.685%

    No Known Activations