© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. MLP Post
    9. 0-MLP-SM
    11. 30980
    Prev
    Next
    INDEX
    Explanations

    words related to the concept of quenching thirst or hydration

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__mlp-sm_processed/0-mlp-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.0.hook_mlp_out
    Hook Layer
    0
    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
    164
    +0.18
    1.0%
    172
    +0.15
    0.8%
    488
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    488
    +0.18
    0.03
    172
    +0.15
    0.02
    489
    +0.14
    0.02
    Negative Logits
    ľ
    -2.78
    -2.72
    ↵
    -2.72
    č↵č↵
    -2.72
    č↵
    -2.72
    ↵
    -2.72
    -2.72
    <|outofrange|>
    -2.72
            ↵
    -2.72
    -2.72
    POSITIVE LOGITS
     aloud
    1.60
    enches
    1.59
    erals
    1.58
    lee
    1.55
    acks
    1.51
    !),
    1.50
    itting
    1.49
    entin
    1.48
    ’
    1.48
    ppo
    1.46
    Activations Density 0.010%

    No Known Activations