© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 4-RES-SM
    11. 10840
    Prev
    Next
    INDEX
    Explanations

    terms related to specific cell types and their characteristics

    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
    348
    +0.18
    1.0%
    475
    +0.12
    0.7%
    18
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    47
    +0.18
    0.01
    18
    +0.12
    0.02
    30
    +0.11
    0.01
    Negative Logits
    dered
    -1.76
     SPSS
    -1.73
     aloud
    -1.70
    fully
    -1.62
    >=
    -1.60
    ably
    -1.60
    IEEE
    -1.55
     offender
    -1.52
     IEEE
    -1.51
    llll
    -1.48
    POSITIVE LOGITS
    µ
    3.60
    ĸ
    3.51
    ¦
    3.43
    ¿
    3.43
    ł
    3.42
    °
    3.33
    ĺ
    3.31
    ·¸
    3.26
    ŀ
    3.25
    ij
    3.24
    Activations Density 0.046%

    No Known Activations