© 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
    2. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    3. GPT2-Small
    4. Residual Post - 32k
    5. 10-RES_POST_32K-OAI
    6. 29545
    Prev
    Next
    INDEX
    Explanations

    terms related to introduction and reintroduction of concepts or entities, particularly in a historical or societal context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_10.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    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
    471
    +0.12
    0.5%
    495
    +0.10
    0.5%
    714
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.12
    0.04
    746
    +0.10
    0.02
    232
    +0.10
    0.03
    Negative Logits
     negatively
    -0.76
    RAW
    -0.71
     firsthand
    -0.70
     captcha
    -0.67
    IFF
    -0.65
     cha
    -0.61
     electroly
    -0.60
    ores
    -0.58
     Bulletin
    -0.58
    ore
    -0.58
    POSITIVE LOGITS
    ctory
    1.17
    cing
    1.09
    cent
    1.05
    fusc
    1.00
    ces
    0.99
    ments
    0.90
    ced
    0.89
    TAIN
    0.86
    ctions
    0.85
    figure
    0.84
    Activations Density 0.184%

    No Known Activations