© 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. 29923
    Prev
    Next
    INDEX
    Explanations

    specific identifiers and license information related to software licenses

    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
    186
    +0.12
    0.7%
    399
    +0.12
    0.7%
    271
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    399
    +0.12
    0.03
    230
    +0.12
    0.03
    74
    +0.12
    0.03
    Negative Logits
     âĢ¦
    -1.46
    `)
    -1.38
    ,’
    -1.36
    Updated
    -1.36
     “
    -1.36
     temperatures
    -1.35
     <-
    -1.32
    while
    -1.32
     Scientists
    -1.31
    âĢ¦..
    -1.30
    POSITIVE LOGITS
    Ĥ¬
    1.73
    ¼
    1.66
    ı
    1.64
    ible
    1.60
    º
    1.56
    Ĵ
    1.51
     repository
    1.48
    opf
    1.46
     unpublished
    1.45
    ¹
    1.44
    Activations Density 0.470%

    No Known Activations