© 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
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. Residuals
    5. 2-RES-SM
    6. 28711
    Prev
    Next
    INDEX
    Explanations

    expressions related to frustration or strong emotions

    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
    479
    +0.21
    1.3%
    423
    +0.16
    1.0%
    204
    +0.16
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    479
    +0.21
    0.12
    259
    +0.16
    0.06
    231
    +0.16
    0.13
    Negative Logits
     sake
    -1.60
     better
    -1.50
     longer
    -1.49
    harmonic
    -1.43
     foreseeable
    -1.39
    astom
    -1.37
    iplex
    -1.34
     purposes
    -1.32
    igent
    -1.32
    sciously
    -1.32
    POSITIVE LOGITS
    §
    3.75
    ¿½
    3.67
    Ļª
    3.63
    Ģ
    3.48
    ĸ
    3.46
    ¦
    3.42
    ĨĴ
    3.42
    ª
    3.39
    Ŀ
    3.36
    Ļ
    3.36
    Activations Density 4.058%

    No Known Activations