© 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. 11379
    Prev
    Next
    INDEX
    Explanations

    aspects related to institutional or organizational collaborations and structural details

    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
    320
    +0.16
    0.9%
    263
    +0.14
    0.8%
    326
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.16
    0.04
    448
    +0.14
    0.03
    252
    +0.12
    0.07
    Negative Logits
    ptions
    -1.67
    ctors
    -1.65
    strings
    -1.58
    ignon
    -1.58
    appings
    -1.57
    initions
    -1.51
     punches
    -1.51
    swers
    -1.47
    annon
    -1.44
    examination
    -1.42
    POSITIVE LOGITS
    )].
    1.38
     ^
    1.36
     AST
    1.34
     Educational
    1.34
    upal
    1.31
     ^[@
    1.30
     supervised
    1.30
     Professional
    1.29
    ĥ½
    1.29
     daughter
    1.29
    Activations Density 1.394%

    No Known Activations