© 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. 10293
    Prev
    Next
    INDEX
    Explanations

    phrases related to the concept of remaining or not leaving a place

    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
    335
    +0.15
    0.8%
    115
    +0.13
    0.7%
    214
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    335
    +0.15
    0.03
    115
    +0.13
    0.02
    361
    +0.10
    0.02
    Negative Logits
    ľ
    -3.48
    ŀ
    -3.39
    ĵ
    -3.38
    Ŀ
    -3.24
    Ħ
    -3.16
    «
    -3.11
    ½
    -3.04
    ĻĤ
    -2.96
    ®
    -2.92
    ¾
    -2.86
    POSITIVE LOGITS
     away
    1.88
     indoors
    1.75
    enthal
    1.68
     alive
    1.61
    olu
    1.58
    over
    1.58
    ders
    1.55
     tuned
    1.49
     stranded
    1.49
    ol
    1.49
    Activations Density 0.151%

    No Known Activations