© 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. 5288
    Prev
    Next
    INDEX
    Explanations

    phrases that emphasize personal consideration or assessment

    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
    71
    +0.18
    0.9%
    270
    +0.10
    0.5%
    155
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    528
    +0.18
    0.03
    362
    +0.10
    0.04
    270
    +0.10
    0.01
    Negative Logits
    ixie
    -0.76
    hent
    -0.70
    enne
    -0.68
    Sys
    -0.64
    lyn
    -0.62
    kered
    -0.60
    oln
    -0.59
    apeake
    -0.58
    erno
    -0.58
    mbuds
    -0.57
    POSITIVE LOGITS
     ramifications
    1.03
     implications
    1.02
     possibilities
    0.91
     alternatives
    0.89
     possibility
    0.88
     probabilities
    0.88
     advis
    0.86
     consequences
    0.84
     options
    0.84
     contribution
    0.84
    Activations Density 0.606%

    No Known Activations