© 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. 10760
    Prev
    Next
    INDEX
    Explanations

    words that describe changes, alterations, or transformations in objects or systems

    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
    156
    +0.20
    1.1%
    25
    +0.12
    0.7%
    391
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    304
    +0.20
    0.07
    149
    +0.12
    0.06
    5
    +0.11
    0.03
    Negative Logits
    -7.10
              ↵
    -7.10
    ↵
    -7.10
    č↵
    -7.10
    -7.10
    <|outofrange|>
    -7.10
    ↵
    -7.10
    -7.10
    ↵
    -7.10
    ↵  ↵
    -7.10
    POSITIVE LOGITS
     into
    2.87
     by
    2.64
     onto
    2.46
     therefrom
    2.41
     within
    2.25
     thereafter
    2.23
     pursuant
    2.05
     therein
    2.04
     with
    2.02
     upon
    1.95
    Activations Density 0.762%

    No Known Activations