© 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. 5303
    Prev
    Next
    INDEX
    Explanations

    words and phrases related to financial and legal terminology

    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
    459
    +0.13
    0.6%
    389
    +0.11
    0.5%
    175
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    679
    +0.13
    0.13
    189
    +0.11
    0.10
    389
    +0.10
    0.08
    Negative Logits
    laughs
    -0.74
    !.
    -0.72
    .
    -0.71
    CI
    -0.70
    +.
    -0.67
    $.
    -0.63
    ;
    -0.63
    rather
    -0.62
    Adds
    -0.61
    .;
    -0.60
    POSITIVE LOGITS
     are
    1.40
     were
    1.29
     aren
    1.28
     weren
    1.24
     comprise
    1.13
     remain
    1.12
     seem
    1.11
     have
    1.10
     cannot
    1.09
     consist
    1.08
    Activations Density 1.094%

    No Known Activations