© 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. 27389
    Prev
    Next
    INDEX
    Explanations

    references to unspecified objects or entities

    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
    434
    +0.13
    0.7%
    93
    +0.12
    0.7%
    241
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    386
    +0.13
    0.02
    59
    +0.12
    0.02
    445
    +0.11
    0.01
    Negative Logits
    t
    -1.76
     Caption
    -1.66
    s
    -1.50
    local
    -1.47
     msgid
    -1.46
    tin
    -1.44
    Willi
    -1.43
    igraph
    -1.42
    temporal
    -1.41
     \%
    -1.40
    POSITIVE LOGITS
    »
    2.68
    ½
    2.63
    IJ
    2.33
    Ĭ
    2.30
    ®
    2.30
    ľĵ
    2.29
    Ŀ
    2.26
    ·¸
    2.24
    ¤
    2.17
    ľ
    2.15
    Activations Density 0.108%

    No Known Activations