© 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
    3. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 8-RES_POST_32K-OAI
    11. 5356
    Prev
    Next
    INDEX
    Explanations

    the word "surely" and its variations, which indicate a sense of certainty or assumption

    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
    187
    +0.10
    0.5%
    526
    +0.09
    0.4%
    636
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    8
    +0.10
    0.02
    718
    +0.09
    0.02
    151
    +0.09
    0.02
    Negative Logits
    ocene
    -0.77
    iatus
    -0.71
    NING
    -0.70
    isphere
    -0.65
    pillar
    -0.64
    pill
    -0.63
    inse
    -0.63
    matic
    -0.63
    nels
    -0.63
    lvl
    -0.63
    POSITIVE LOGITS
     someday
    0.83
    some
    0.70
    saf
    0.67
    ortment
    0.64
     await
    0.63
     awaits
    0.63
    ä¹
    0.62
     perished
    0.61
    lessly
    0.60
    éĹ
    0.59
    Activations Density 0.053%

    No Known Activations