© 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. 2-RES-SM
    11. 10251
    Prev
    Next
    INDEX
    Explanations

    actions or phrases related to throwing or discarding items

    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
    127
    +0.15
    0.8%
    148
    +0.12
    0.6%
    258
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    148
    +0.15
    0.01
    62
    +0.12
    0.02
    127
    +0.11
    0.03
    Negative Logits
    aliana
    -1.93
    ial
    -1.54
     stake
    -1.44
    ately
    -1.41
     Sector
    -1.39
    rically
    -1.39
    artments
    -1.39
     Available
    -1.38
     exists
    -1.38
    esa
    -1.38
    POSITIVE LOGITS
    away
    2.21
    aways
    2.02
    tails
    1.93
     away
    1.76
     err
    1.65
    offs
    1.64
    off
    1.58
    backs
    1.53
     bombs
    1.52
    lasses
    1.51
    Activations Density 0.255%

    No Known Activations