© 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. 5291
    Prev
    Next
    INDEX
    Explanations

    entities and key players involved in legal cases and investigations

    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
    373
    +0.14
    0.6%
    31
    +0.12
    0.5%
    79
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.14
    0.19
    146
    +0.12
    0.09
    224
    +0.11
    0.10
    Negative Logits
    liness
    -0.80
    peak
    -0.78
     smiles
    -0.75
     proverb
    -0.74
     cheers
    -0.68
     slang
    -0.68
     laughs
    -0.66
     magnification
    -0.66
    iquette
    -0.65
     roar
    -0.65
    POSITIVE LOGITS
     defendants
    1.00
     violated
    0.98
     unlawfully
    0.92
     plaintiffs
    0.91
     alleging
    0.90
     liable
    0.88
     violates
    0.83
     criminally
    0.82
     violating
    0.80
     viol
    0.80
    Activations Density 3.294%

    No Known Activations