© 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. 22492
    Prev
    Next
    INDEX
    Explanations

    instances of "more information" or similar phrases indicating the availability of additional details

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Neuron Alignment
    Index
    Value
    % of L₁
    156
    +0.17
    1.0%
    435
    +0.16
    0.9%
    211
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    59
    +0.17
    -0.03
    418
    +0.16
    -0.04
    489
    +0.12
    0.00
    Negative Logits
    noreply
    -1.63
    aeda
    -1.58
    iator
    -1.54
    isenberg
    -1.46
    irds
    -1.42
     why
    -1.42
    ↵³³
    -1.42
    PHA
    -1.40
    icago
    -1.39
    asone
    -1.34
    POSITIVE LOGITS
    Ħ
    3.01
    Ĥ
    2.76
    ¾
    2.69
    Ĺ
    2.58
    º
    2.57
    £
    2.54
    ij
    2.43
    ±
    2.38
    ª
    2.35
    ´
    2.33
    Activations Density 2.498%

    No Known Activations