© 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. 1-RES-SM
    11. 25104
    Prev
    Next
    INDEX
    Explanations

    references to global themes and issues

    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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    156
    +0.21
    1.2%
    29
    +0.15
    0.8%
    119
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    29
    +0.21
    0.03
    429
    +0.15
    0.03
    36
    +0.11
    0.03
    Negative Logits
    rir
    -1.69
    ository
    -1.67
    lessly
    -1.66
     arbitrary
    -1.60
     sacrifice
    -1.59
    ellow
    -1.55
    hler
    -1.55
    ching
    -1.53
    ck
    -1.48
    ktion
    -1.44
    POSITIVE LOGITS
    ķ
    2.59
    Ł
    2.46
    Ĺ
    2.36
    ·¸
    2.33
    °
    2.25
    Ħ
    2.18
    ĺ
    2.09
    £
    2.05
    »¿
    2.05
    ĵ
    2.03
    Activations Density 0.054%

    No Known Activations