© 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. 5286
    Prev
    Next
    INDEX
    Explanations

    key terms related to products, models, and significant concepts in technology and media

    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
    288
    +0.12
    0.6%
    408
    +0.10
    0.5%
    44
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.12
    0.14
    408
    +0.10
    0.10
    21
    +0.10
    0.05
    Negative Logits
    tics
    -1.08
    ounds
    -0.89
    cale
    -0.87
    arus
    -0.87
    ledge
    -0.81
    aves
    -0.80
    ascript
    -0.78
    aving
    -0.77
    renches
    -0.76
    ucket
    -0.76
    POSITIVE LOGITS
     employer
    0.82
     batch
    0.81
     piece
    0.81
     entity
    0.80
     organism
    0.80
     item
    0.79
     product
    0.78
     article
    0.78
     audience
    0.78
     letter
    0.77
    Activations Density 1.420%

    No Known Activations