© 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. 5258
    Prev
    Next
    INDEX
    Explanations

    phrases indicating division or destruction

    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
    314
    +0.11
    0.5%
    591
    +0.10
    0.4%
    361
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    608
    +0.11
    0.06
    526
    +0.10
    0.00
    113
    +0.10
    0.04
    Negative Logits
    eton
    -0.75
     Anthem
    -0.67
     risky
    -0.67
    eus
    -0.67
     Warn
    -0.66
     IPM
    -0.64
     Mutual
    -0.63
     Situation
    -0.63
     Innov
    -0.63
     facult
    -0.61
    POSITIVE LOGITS
     shred
    1.14
     pieces
    1.04
     shape
    0.90
     fragments
    0.88
     atoms
    0.82
     bones
    0.81
     bits
    0.81
     shapes
    0.80
     cubes
    0.80
     Pieces
    0.79
    Activations Density 0.554%

    No Known Activations