© 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. 4697
    Prev
    Next
    INDEX
    Explanations

    intensifiers that convey strong emotions or states

    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
    233
    +0.10
    0.4%
    608
    +0.09
    0.4%
    393
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    227
    +0.10
    0.02
    284
    +0.09
    0.03
    348
    +0.09
    0.03
    Negative Logits
    ascript
    -0.74
    enment
    -0.68
    idas
    -0.64
    ando
    -0.60
    achus
    -0.60
     Incident
    -0.58
    afety
    -0.58
    itures
    -0.58
    anchester
    -0.57
    acas
    -0.57
    POSITIVE LOGITS
     WIN
    0.59
     istg
    0.58
     fucked
    0.55
    WHERE
    0.55
     lett
    0.54
     RU
    0.54
    llah
    0.54
    ibur
    0.54
    ittle
    0.54
    mean
    0.53
    Activations Density 0.176%

    No Known Activations