© 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
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 2-RES-SM
    11. 11317
    Prev
    Next
    INDEX
    Explanations

    numeric identifiers and codes

    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
    186
    +0.18
    1.1%
    293
    +0.13
    0.8%
    6
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    6
    +0.18
    0.17
    239
    +0.13
    0.19
    243
    +0.13
    0.18
    Negative Logits
    onucle
    -1.30
     loosely
    -1.29
    )?
    -1.20
     exclusively
    -1.20
    semble
    -1.19
     Register
    -1.18
    atan
    -1.17
    anne
    -1.16
    aya
    -1.16
     Broadcasting
    -1.15
    POSITIVE LOGITS
    §
    2.08
    ¥
    2.04
    ²
    1.82
    ½
    1.77
    Ĺ
    1.73
    Į
    1.73
    Ł
    1.72
    :`
    1.72
    ¡
    1.66
    ª
    1.64
    Activations Density 6.584%

    No Known Activations