© 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. 5257
    Prev
    Next
    INDEX
    Explanations

    phrases related to managing situations or outcomes

    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
    568
    +0.10
    0.5%
    73
    +0.09
    0.4%
    665
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    665
    +0.10
    0.07
    70
    +0.09
    0.05
    469
    +0.09
    0.03
    Negative Logits
    imilar
    -0.63
    alky
    -0.63
    doi
    -0.61
    ASC
    -0.59
     reversal
    -0.58
     Cosponsors
    -0.58
    ãĤ·ãĥ£
    -0.58
    quickShipAvailable
    -0.56
    similar
    -0.55
    acist
    -0.55
    POSITIVE LOGITS
     freely
    1.03
     unrestricted
    0.78
     unchecked
    0.78
     theirs
    0.70
     safely
    0.66
    ique
    0.64
     peacefully
    0.64
     free
    0.64
     loose
    0.64
     bask
    0.64
    Activations Density 0.521%

    No Known Activations