© 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. 29900
    Prev
    Next
    INDEX
    Explanations

    descriptive adjectives related to clothing and textiles

    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
    153
    +0.15
    0.8%
    198
    +0.12
    0.7%
    32
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    407
    +0.15
    0.13
    16
    +0.12
    0.09
    14
    +0.11
    0.15
    Negative Logits
    ł
    -2.19
    £
    -2.16
    ŀ
    -2.09
    overnment
    -2.05
    ±
    -1.96
    Ĩ
    -1.95
    Īĺ
    -1.85
     election
    -1.84
     hearings
    -1.77
    ļ
    -1.75
    POSITIVE LOGITS
     comfort
    1.99
     sleek
    1.87
     stylish
    1.86
     styling
    1.86
     versatility
    1.79
     decorative
    1.75
     customize
    1.73
    lux
    1.72
     beauty
    1.70
     durability
    1.65
    Activations Density 6.328%

    No Known Activations