© 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. 10766
    Prev
    Next
    INDEX
    Explanations

    the word "Bl" or variations of it within the text

    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
    477
    +0.18
    1.0%
    156
    +0.13
    0.7%
    185
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    477
    +0.18
    0.05
    185
    +0.13
    0.02
    59
    +0.12
    0.02
    Negative Logits
    unnumbered
    -1.68
    PLIED
    -1.66
    ria
    -1.61
    ocratic
    -1.58
     right
    -1.57
    ional
    -1.56
    cias
    -1.50
    »¿
    -1.50
    ylene
    -1.47
     minist
    -1.44
    POSITIVE LOGITS
    ings
    2.29
    uetooth
    2.18
    acks
    1.89
    ishes
    1.79
    uffs
    1.76
    itz
    1.75
    ouin
    1.75
    enders
    1.74
    izzard
    1.74
    azer
    1.69
    Activations Density 0.118%

    No Known Activations