© 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. 1-RES-SM
    11. 26624
    Prev
    Next
    INDEX
    Explanations

    symbols and characters from various languages or scripts

    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
    410
    +0.25
    1.4%
    369
    +0.22
    1.3%
    156
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    111
    +0.25
    0.10
    410
    +0.22
    0.07
    271
    +0.11
    0.01
    Negative Logits
    ETHOD
    -1.75
    izable
    -1.69
    ailable
    -1.67
    ivas
    -1.63
    aset
    -1.59
    ample
    -1.58
    ibility
    -1.58
    ength
    -1.54
    axel
    -1.52
    effectiveness
    -1.51
    POSITIVE LOGITS
    ²
    3.80
    Ĵ
    3.80
    ī
    3.65
    ¼
    3.64
    «
    3.50
    ģ
    3.49
    Į
    3.40
    ®
    3.38
    Ŀ
    3.35
    ŀ
    3.33
    Activations Density 0.595%

    No Known Activations