© 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. 9559
    Prev
    Next
    INDEX
    Explanations

    references to personal relationships and emotional connections

    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
    293
    +0.14
    0.8%
    25
    +0.12
    0.6%
    149
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.14
    0.14
    170
    +0.12
    0.11
    293
    +0.11
    0.11
    Negative Logits
    esto
    -1.61
    rapeutic
    -1.57
     Chern
    -1.47
    Ĥ
    -1.39
    estic
    -1.39
    Ī
    -1.38
     variance
    -1.35
    ¦
    -1.34
    Ĩ
    -1.32
     regards
    -1.31
    POSITIVE LOGITS
    blr
    1.59
    ilate
    1.53
    /~
    1.52
     awhile
    1.51
    ():
    1.50
    upal
    1.50
     yourself
    1.46
    ocks
    1.42
     <-
    1.42
     INTO
    1.41
    Activations Density 2.067%

    No Known Activations