© 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. 10465
    Prev
    Next
    INDEX
    Explanations

    instances of the word 'very' followed by an adjective

    oai_token-act-pair · gpt-3.5-turbo

    intensifiers that describe qualities of objects or experiences

    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
    anwhile
    -0.84
     elsewhere
    -0.76
    onto
    -0.75
    omever
    -0.73
     policies
    -0.73
     quotas
    -0.72
    Officials
    -0.71
    pta
    -0.71
    livion
    -0.71
    ['
    -0.71
    POSITIVE LOGITS
     informative
    1.08
     customizable
    1.06
     underrated
    0.99
     playable
    0.99
     versatile
    0.94
     simplified
    0.92
     beginner
    0.90
     unique
    0.89
     concise
    0.86
     readable
    0.86
    Activations Density 0.419%

    No Known Activations