© 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. 5398
    Prev
    Next
    INDEX
    Explanations

    instances of the word "exp" in various contexts

    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
    208
    +0.10
    0.5%
    159
    +0.09
    0.4%
    273
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    208
    +0.10
    0.02
    670
    +0.09
    0.02
    273
    +0.09
    0.02
    Negative Logits
    ball
    -0.79
    footed
    -0.75
    balls
    -0.74
    BOOK
    -0.67
    crow
    -0.67
    pants
    -0.65
    mates
    -0.64
     thumbs
    -0.64
    DOWN
    -0.64
    BALL
    -0.63
    POSITIVE LOGITS
    onential
    1.52
    ansion
    1.46
    onent
    1.41
    ulsion
    1.40
    anded
    1.39
    osed
    1.34
    anding
    1.33
    ression
    1.29
    anse
    1.28
    orters
    1.21
    Activations Density 0.028%

    No Known Activations