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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 8-RES_POST_32K-OAI
    11. 5584
    Prev
    Next
    INDEX
    Explanations

    informal forms of address or interjections commonly used in casual conversation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_8.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.8.hook_resid_post
    Hook Layer
    8
    Architecture
    standard
    Context Size
    64
    Dataset
    Skylion007/openwebtext
    Activation Function
    topk
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    420
    +0.13
    0.6%
    316
    +0.11
    0.5%
    54
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    696
    +0.13
    0.03
    193
    +0.11
    0.03
    293
    +0.10
    0.02
    Negative Logits
    Balt
    -0.79
     Canaver
    -0.74
    ãĤ¹ãĥĪ
    -0.72
    ł
    -0.71
    ENG
    -0.68
    ³
    -0.68
    ãĥ¼ãĥ«
    -0.67
    ãĤ»
    -0.67
    IRC
    -0.66
    ħ
    -0.65
    POSITIVE LOGITS
    selves
    1.50
    self
    1.21
    're
    0.92
     guys
    0.91
    'll
    0.91
     know
    0.88
     kid
    0.82
     boys
    0.82
     fuckin
    0.81
     gotta
    0.80
    Activations Density 0.198%

    No Known Activations