© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 1-RES-SM
    11. 29057
    Prev
    Next
    INDEX
    Explanations

    technical specifications and operational limitations related to a system or service

    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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    156
    +0.39
    2.6%
    111
    +0.24
    1.6%
    17
    +0.17
    1.1%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    111
    +0.39
    0.42
    271
    +0.24
    0.20
    156
    +0.17
    0.20
    Negative Logits
     surfaces
    -1.53
    asone
    -1.39
    ESULT
    -1.38
     determinants
    -1.36
    1451450014514500
    -1.36
    iza
    -1.35
    ocy
    -1.33
    ias
    -1.30
    ppat
    -1.29
    ariat
    -1.28
    POSITIVE LOGITS
    §
    2.07
    ¼
    2.06
    ³
    2.06
    ÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤ
    2.05
    Ĥ
    2.03
    »¿
    2.03
    Ħ
    1.98
    Ļ
    1.98
    Ĺ
    1.97
    ī
    1.95
    Activations Density 0.793%

    No Known Activations