© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 12277
    Prev
    Next
    INDEX
    Explanations

    programming function and method calls, particularly related to variable handling and data manipulation

    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
    292
    +0.13
    0.7%
    281
    +0.11
    0.6%
    144
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    29
    +0.13
    0.03
    188
    +0.11
    0.02
    368
    +0.11
    0.02
    Negative Logits
    hip
    -1.55
    hin
    -1.55
    oning
    -1.51
    olev
    -1.46
    itt
    -1.40
    iana
    -1.39
    achus
    -1.38
     classics
    -1.38
    ians
    -1.36
    optera
    -1.35
    POSITIVE LOGITS
     veins
    1.55
    ovsk
    1.52
     host
    1.41
     generality
    1.34
    vey
    1.32
    acious
    1.30
    dock
    1.25
     presumably
    1.25
    fn
    1.20
    lette
    1.20
    Activations Density 0.096%

    No Known Activations