© 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. 10-RES-JB
    11. 7905
    Prev
    Next
    INDEX
    Explanations

    Cyrillic characters, particularly the letter "е"

    oai_token-act-pair · gpt-3.5-turbo

    character strings that resemble foreign or coded text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ailability
    -0.81
    FUL
    -0.67
    cliffe
    -0.67
    mods
    -0.66
    ĵĺ
    -0.64
    iqueness
    -0.64
    INGTON
    -0.62
    aughs
    -0.61
     onions
    -0.59
    chel
    -0.59
    POSITIVE LOGITS
    е
    0.78
    ©¶æ
    0.73
    lict
    0.72
    ulate
    0.69
    Mine
    0.68
     Dum
    0.68
    Ñ
    0.67
    к
    0.66
    ++)
    0.65
    Ñı
    0.65
    Activations Density 0.075%

    No Known Activations