© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 3-RES-JB
    11. 6165
    Prev
    Next
    INDEX
    Explanations

    non-English characters and symbols, possibly indicating a preference for a specific language or script

    oai_token-act-pair · gpt-3.5-turbo

    characters or symbols that predominantly appear in different languages, particularly those with diacritics or special characters

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    este
    -0.84
    anon
    -0.81
    mast
    -0.79
    otle
    -0.79
    PT
    -0.79
    ebus
    -0.78
    Maker
    -0.75
    eled
    -0.74
    ened
    -0.74
    eele
    -0.74
    POSITIVE LOGITS
    «ĺ
    0.97
    ¬¼
    0.91
    ĨĴ
    0.76
    м
    0.74
     breathing
    0.69
    ŃĶ
    0.69
    Ñı
    0.67
     arming
    0.66
    lict
    0.65
     tending
    0.63
    Activations Density 0.011%

    No Known Activations