© 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. 28702
    Prev
    Next
    INDEX
    Explanations

    terms related to live broadcasts or streaming events

    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
    477
    +0.13
    0.7%
    214
    +0.12
    0.7%
    484
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    29
    +0.13
    0.01
    214
    +0.12
    0.02
    477
    +0.12
    0.03
    Negative Logits
    Ń
    -3.07
    ¯
    -3.04
    ·¸
    -2.95
    ľĵ
    -2.89
    ľ
    -2.75
    Ĺ
    -2.75
    Ĥ
    -2.72
    §
    -2.70
    Ŀ
    -2.65
    ¤
    -2.58
    POSITIVE LOGITS
     footage
    1.89
     broadcasts
    1.82
    cd
    1.75
    cats
    1.72
    !,
    1.72
    cat
    1.68
    arts
    1.67
     broadcast
    1.65
    journal
    1.63
    fires
    1.61
    Activations Density 0.107%

    No Known Activations