© 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. 5501
    Prev
    Next
    INDEX
    Explanations

    references to characters and their relationships in an action context

    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
    516
    +0.12
    0.5%
    430
    +0.12
    0.5%
    185
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    139
    +0.12
    0.13
    354
    +0.12
    0.12
    430
    +0.10
    0.13
    Negative Logits
     Pigs
    -0.73
     Favor
    -0.69
    hua
    -0.67
    Num
    -0.63
     Signature
    -0.63
     Bloomberg
    -0.62
     Illum
    -0.61
     Soda
    -0.61
     Pyramid
    -0.61
    Regular
    -0.60
    POSITIVE LOGITS
     eventually
    1.05
     promptly
    1.05
     reluctantly
    1.04
     soon
    0.97
     quickly
    0.97
     regained
    0.94
     finally
    0.92
     immediately
    0.91
     swiftly
    0.91
     subsequently
    0.90
    Activations Density 1.621%

    No Known Activations