© 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. 12592
    Prev
    Next
    INDEX
    Explanations

    expressions of warranties and conditions in licenses

    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
    47
    +0.12
    0.7%
    320
    +0.11
    0.6%
    463
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    463
    +0.12
    0.02
    222
    +0.11
    0.02
    487
    +0.11
    0.01
    Negative Logits
    ľĵ
    -1.79
    apper
    -1.56
    ennis
    -1.47
    unge
    -1.44
    atom
    -1.44
    allas
    -1.44
    '))
    -1.40
    opal
    -1.39
    cyst
    -1.39
    aster
    -1.36
    POSITIVE LOGITS
     constants
    1.48
     WARRANTIES
    1.40
    written
    1.36
    minded
    1.34
     degrees
    1.32
    ervice
    1.32
     oneself
    1.31
    eless
    1.31
    iousness
    1.27
    itud
    1.27
    Activations Density 0.037%

    No Known Activations