© 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. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 4-RES-JB
    11. 10916
    Prev
    Next
    INDEX
    Explanations

    Asian characters and symbol patterns

    oai_token-act-pair · gpt-3.5-turbo

    non-Latin characters or symbols, indicating a focus on diverse language representations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 4-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.4.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    4
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     subord
    -0.78
    enegger
    -0.75
    emale
    -0.75
     Seym
    -0.73
    ription
    -0.72
    raints
    -0.72
    schild
    -0.72
     horizont
    -0.71
    pheus
    -0.70
    espie
    -0.69
    POSITIVE LOGITS
    ħ
    0.80
    Ī
    0.79
    ĺ
    0.77
    ĥ
    0.75
    RAM
    0.73
    à¨
    0.72
    BILITIES
    0.72
    Ĥª
    0.72
    Ļ
    0.71
    ãĥŃ
    0.71
    Activations Density 0.037%

    No Known Activations