© 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
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 4-RES-SM
    11. 11852
    Prev
    Next
    INDEX
    Explanations

    terms associated with the act of supplying or giving something

    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
    443
    +0.13
    0.8%
    290
    +0.12
    0.7%
    217
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    290
    +0.13
    0.01
    443
    +0.12
    0.01
    157
    +0.10
    0.01
    Negative Logits
    lotte
    -1.39
     decir
    -1.36
    cheon
    -1.33
    ainen
    -1.31
    raits
    -1.31
    obile
    -1.30
    hesis
    -1.30
    ribe
    -1.29
    ola
    -1.26
    rap
    -1.25
    POSITIVE LOGITS
    §
    1.91
    ³
    1.87
    ĻĤ
    1.87
    ¥
    1.83
    ¬
    1.80
    ¶
    1.79
    Ĵ
    1.75
    ¤
    1.66
    ļ
    1.65
    º
    1.62
    Activations Density 0.036%

    No Known Activations