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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 28707
    Prev
    Next
    INDEX
    Explanations

    numeric values and timestamps

    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
    497
    +0.13
    0.7%
    12
    +0.13
    0.7%
    24
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    497
    +0.13
    0.04
    12
    +0.13
    0.09
    17
    +0.12
    -0.01
    Negative Logits
     revived
    -1.42
     monopoly
    -1.38
     application
    -1.33
     doctrine
    -1.31
     synonymous
    -1.31
     discovery
    -1.30
     consortium
    -1.30
     Johannes
    -1.28
    esters
    -1.28
     competence
    -1.27
    POSITIVE LOGITS
    ·¸
    3.97
    Ŀ
    3.77
    IJ
    3.70
    Īĺ
    3.65
    ¤
    3.62
    ĥ½
    3.61
    Ĵ
    3.50
    ±
    3.50
    £
    3.44
    ĻĤ
    3.36
    Activations Density 2.306%

    No Known Activations