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

    Neuronpedia

    Jacobian LensNEW
    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. 8-RES-JB
    11. 19338
    Prev
    Next
    INDEX
    Explanations

    names or terms appearing to be related to political figures or entities

    oai_token-act-pair · gpt-3.5-turbo

    the word "ats" and its variations, indicating a focus on a specific term or abbreviation related to a particular context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ©¶æ
    -0.67
    ously
    -0.62
     sclerosis
    -0.62
    shire
    -0.59
     Vaugh
    -0.59
    SOURCE
    -0.58
    Blade
    -0.58
    ADE
    -0.56
     Sparkle
    -0.54
     fracturing
    -0.54
    POSITIVE LOGITS
    terson
    1.09
    rix
    1.06
    rice
    1.06
    ting
    1.01
    ches
    1.00
    tered
    0.99
    ICAN
    0.99
    herer
    0.98
    hematic
    0.98
    chers
    0.97
    Activations Density 0.059%

    No Known Activations