© 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. 14003
    Prev
    Next
    INDEX
    Explanations

    phrases related to thermal properties and temperature management in various contexts

    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
    47
    +0.19
    1.1%
    231
    +0.12
    0.7%
    259
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    47
    +0.19
    0.10
    259
    +0.12
    0.03
    231
    +0.11
    0.07
    Negative Logits
     puzzle
    -1.80
     tackles
    -1.68
     appeals
    -1.65
     looks
    -1.63
     stats
    -1.59
     deals
    -1.58
     lists
    -1.58
     Ajax
    -1.57
    iary
    -1.55
     rules
    -1.51
    POSITIVE LOGITS
     temperatures
    2.34
     humidity
    2.29
    mosp
    1.95
     substrate
    1.94
     precipitation
    1.90
    sein
    1.87
     temperature
    1.85
     humid
    1.82
    olerance
    1.80
     pH
    1.72
    Activations Density 1.703%

    No Known Activations