© 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. 2-RES-SM
    11. 27076
    Prev
    Next
    INDEX
    Explanations

    technical terms and error messages related to image processing

    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/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
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    501
    +0.15
    0.9%
    261
    +0.12
    0.7%
    497
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    501
    +0.15
    0.03
    445
    +0.12
    0.02
    497
    +0.12
    0.02
    Negative Logits
     focused
    -1.50
     track
    -1.49
    ellants
    -1.42
     consists
    -1.40
    thor
    -1.39
    )](#
    -1.36
     path
    -1.35
     service
    -1.35
     diagnoses
    -1.35
     experienced
    -1.34
    POSITIVE LOGITS
    Īĺ
    2.96
    ł
    2.50
    Ĩ
    2.49
    ¹
    2.39
    Ļ
    2.32
    ¥
    2.32
    ĺ
    2.28
    ±
    2.24
    ħ
    2.24
    ¦
    2.24
    Activations Density 0.224%

    No Known Activations