© 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. 4870
    Prev
    Next
    INDEX
    Explanations

    specific names or terms related to literature and entertainment

    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
    711
    +0.10
    0.5%
    136
    +0.09
    0.4%
    697
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.10
    0.17
    711
    +0.09
    0.04
    461
    +0.09
    0.06
    Negative Logits
     alike
    -0.63
    asca
    -0.58
     Silver
    -0.55
    senal
    -0.55
     pigs
    -0.54
     Cub
    -0.53
     Mug
    -0.50
    othermal
    -0.49
     abs
    -0.48
     gulf
    -0.48
    POSITIVE LOGITS
    and
    1.00
    &
    0.95
     &
    0.95
     AND
    0.94
    /
    0.89
    /
    0.88
     and
    0.86
    /+
    0.81
     or
    0.81
    ands
    0.73
    Activations Density 4.745%

    No Known Activations