© 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. 23737
    Prev
    Next
    INDEX
    Explanations

    character sequences that may not align with text in the document

    oai_token-act-pair · gpt-3.5-turbo

    sequences of special characters and specific terms related to individuals

    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
    ffield
    -0.87
    tha
    -0.85
    meet
    -0.79
    roots
    -0.78
    entimes
    -0.78
    resa
    -0.75
    aired
    -0.73
    iser
    -0.73
    uses
    -0.71
    folio
    -0.71
    POSITIVE LOGITS
    íķ
    0.86
     vinegar
    0.80
    VERTISEMENT
    0.74
     ì
    0.70
     Coke
    0.70
    ì
    0.70
    ëĭ
    0.68
     Kore
    0.67
    ê
    0.67
    hal
    0.66
    Activations Density 0.025%

    No Known Activations