© 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. 2-RES-SM
    6. 30006
    Prev
    Next
    INDEX
    Explanations

    references to being named or designated in a notable context

    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
    236
    +0.12
    0.7%
    330
    +0.11
    0.6%
    150
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    236
    +0.12
    0.02
    138
    +0.11
    0.02
    29
    +0.11
    0.01
    Negative Logits
    ared
    -2.06
    insic
    -1.82
    arity
    -1.70
    orship
    -1.63
    oretical
    -1.55
    ibraries
    -1.51
    ores
    -1.50
    ory
    -1.50
    anguages
    -1.47
    ferential
    -1.47
    POSITIVE LOGITS
    ł
    1.89
    ¿
    1.83
    ¯
    1.77
    ¼
    1.73
    Ĥ
    1.73
    Ģ
    1.73
    ²
    1.65
    ¨
    1.62
    ¤
    1.56
    Ĭ
    1.52
    Activations Density 0.104%

    No Known Activations