© 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. Residuals
    9. 2-RES-SM
    11. 32383
    Prev
    Next
    INDEX
    Explanations

    terms related to merging and publishing actions

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Neuron Alignment
    Index
    Value
    % of L₁
    412
    +0.19
    1.1%
    439
    +0.13
    0.7%
    68
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    412
    +0.19
    0.03
    239
    +0.13
    0.03
    82
    +0.11
    0.01
    Negative Logits
    Ļ
    -2.40
    Ĩ
    -2.35
    §
    -2.18
    ¯
    -2.09
    ´
    -2.02
    ¹
    -2.02
    ³
    -1.98
    ®
    -1.92
    ĥ½
    -1.90
    liest
    -1.89
    POSITIVE LOGITS
    eson
    1.68
     mode
    1.57
    andum
    1.45
     fields
    1.42
     bases
    1.40
    point
    1.35
    hdad
    1.33
     modes
    1.33
     purposes
    1.31
     Point
    1.31
    Activations Density 0.152%

    No Known Activations