© 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. 5346
    Prev
    Next
    INDEX
    Explanations

    conjunctions and phrases indicating relationships or dependencies

    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
    756
    +0.15
    0.7%
    459
    +0.13
    0.6%
    256
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    705
    +0.15
    0.08
    601
    +0.13
    0.05
    203
    +0.10
    0.06
    Negative Logits
    Anyway
    -0.93
    Yeah
    -0.92
    yeah
    -0.91
    maybe
    -0.91
    Yep
    -0.90
    Maybe
    -0.88
     piss
    -0.88
     freaking
    -0.86
    Pretty
    -0.84
     kinda
    -0.84
    POSITIVE LOGITS
     persons
    0.97
     corrective
    0.91
     deviations
    0.90
     provisions
    0.85
     §§
    0.84
     residual
    0.84
     appellant
    0.83
     substantial
    0.83
     Additional
    0.82
     applic
    0.82
    Activations Density 0.369%

    No Known Activations