© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 4-RES-SM
    11. 13096
    Prev
    Next
    INDEX
    Explanations

    specific names of individuals and groups associated with events or topics

    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/4-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.4.hook_resid_post
    Hook Layer
    4
    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
    348
    +0.17
    1.0%
    445
    +0.12
    0.7%
    177
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    410
    +0.17
    0.07
    98
    +0.12
    0.04
    177
    +0.11
    0.08
    Negative Logits
    ership
    -1.92
    mable
    -1.83
    cology
    -1.73
    /~
    -1.65
    ington
    -1.63
    mers
    -1.58
    iners
    -1.57
    where
    -1.54
    ynaptic
    -1.54
    copyright
    -1.53
    POSITIVE LOGITS
    sson
    2.20
     Sanchez
    2.15
     Andre
    2.11
     Manuel
    2.10
     Martinez
    2.09
     Brun
    2.08
     Antonio
    2.08
     Stefan
    2.06
     Borg
    2.04
     Valent
    2.02
    Activations Density 0.423%

    No Known Activations