© 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. 5424
    Prev
    Next
    INDEX
    Explanations

    occurrences of the letter 'L' 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
    659
    +0.21
    1.0%
    54
    +0.12
    0.6%
    482
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.21
    0.08
    18
    +0.12
    0.06
    103
    +0.10
    0.04
    Negative Logits
     err
    -0.65
     Conquest
    -0.56
     Rebels
    -0.56
     conversion
    -0.56
     moot
    -0.56
     mileage
    -0.55
     tem
    -0.55
     Blizz
    -0.55
     satire
    -0.55
     trivia
    -0.55
    POSITIVE LOGITS
    arger
    1.10
    educ
    1.02
    ymph
    0.98
    umi
    0.98
    oneliness
    0.95
    izzy
    0.94
    ym
    0.93
    yle
    0.92
    ARP
    0.92
    ately
    0.91
    Activations Density 0.060%

    No Known Activations