© 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. 2-MLP-SM
    11. 22638
    Prev
    Next
    INDEX
    Explanations

    terms and concepts related to support, impact, and development in various contexts

    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/2-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.2.hook_mlp_out
    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
    111
    +0.13
    0.7%
    287
    +0.13
    0.7%
    127
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    127
    +0.13
    0.16
    168
    +0.13
    0.13
    120
    +0.11
    0.16
    Negative Logits
    ister
    -1.61
     odds
    -1.52
     eternity
    -1.46
     innings
    -1.43
    aughters
    -1.43
    ald
    -1.42
     wonders
    -1.41
     indication
    -1.40
    cluded
    -1.38
    )?
    -1.37
    POSITIVE LOGITS
    "`
    1.76
    "_
    1.57
    ructure
    1.52
     quo
    1.48
    ``
    1.48
    etine
    1.48
    ruct
    1.48
    liness
    1.43
    ]'
    1.35
    trip
    1.33
    Activations Density 2.134%

    No Known Activations