© 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. 0-RES-JB
    11. 22298
    Prev
    Next
    INDEX
    Explanations

    references to interactive elements or features

    oai_token-act-pair · gpt-3.5-turbo

    references to interactive content and experiences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ries
    -0.73
     Baron
    -0.72
    ried
    -0.67
     Niet
    -0.67
     disqual
    -0.67
     Bulgar
    -0.66
    _{
    -0.65
    irm
    -0.64
    tha
    -0.64
    ieties
    -0.63
    POSITIVE LOGITS
     interactive
    3.49
     Interactive
    2.15
     immersive
    1.56
     interact
    1.45
     infographic
    1.42
     multimedia
    1.35
     downloadable
    1.24
     informative
    1.23
     animated
    1.22
     touchscreen
    1.20
    Activations Density 0.018%

    No Known Activations