© 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. 0-RES-JB
    11. 8410
    Prev
    Next
    INDEX
    Explanations

    professional titles related to photography and cartography

    oai_token-act-pair · gpt-3.5-turbo

    terms related to various types of photographers and biography writers

    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
     suspension
    -0.68
     fun
    -0.63
     March
    -0.63
    susp
    -0.63
     Mut
    -0.62
     unconditional
    -0.62
     Susp
    -0.62
     Tens
    -0.61
     Tun
    -0.60
     null
    -0.60
    POSITIVE LOGITS
    ographer
    4.58
    ographers
    4.10
    ography
    2.73
    ographic
    2.40
    ographies
    2.39
    ographically
    2.39
    ographical
    2.24
    ograph
    2.18
    ograp
    2.14
    ographs
    1.96
    Activations Density 0.010%

    No Known Activations