Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. Residuals
    5. 2-RES-SM
    6. 12139
    Prev
    Next
    INDEX
    Explanations

    identifiers and references in code or markup language

    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
    +0.16
    0.9%
    183
    +0.12
    0.7%
    307
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    497
    +0.16
    0.06
    187
    +0.12
    0.07
    504
    +0.11
    0.05
    Negative Logits
    obl
    -1.56
    sizeof
    -1.45
    pass
    -1.43
    ––
    -1.38
    cedes
    -1.34
    ensa
    -1.32
     enjoys
    -1.32
     retains
    -1.29
    rand
    -1.28
    throws
    -1.26
    POSITIVE LOGITS
    ĻĤ
    2.62
    ł
    2.37
    °
    2.33
    ·¸
    2.31
    ³
    2.23
    ħ
    2.18
    Ŀ
    2.14
    ľ
    2.12
    ¿
    2.11
    ¦
    2.02
    Activations Density 3.581%

    No Known Activations