© 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. 4-RES-SM
    11. 10409
    Prev
    Next
    INDEX
    Explanations

    instances of the word "in" indicating locations or 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/4-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.4.hook_resid_post
    Hook Layer
    4
    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
    136
    +0.15
    0.8%
    448
    +0.12
    0.7%
    478
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    136
    +0.15
    0.00
    166
    +0.12
    0.06
    139
    +0.11
    0.07
    Negative Logits
    Ĥ¬
    -2.69
    º
    -2.59
    Īĺ
    -2.58
    ²
    -2.52
    ¢
    -2.49
    ĻĤ
    -2.38
    ħ
    -2.33
    ģ
    -2.31
    Ĺ
    -2.22
    ķ
    -2.22
    POSITIVE LOGITS
     1997
    2.77
     1966
    2.73
     1996
    2.73
     1993
    2.72
     1964
    2.67
     2001
    2.65
     1977
    2.65
     1991
    2.64
     1974
    2.63
     1987
    2.63
    Activations Density 0.305%

    No Known Activations