© 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. 8-RES-JB
    11. 10099
    Prev
    Next
    INDEX
    Explanations

    adjectives denoting high praise or criticism

    oai_token-act-pair · gpt-3.5-turbo

    adjectives that describe qualities or characteristics

    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
    ecause
    -0.85
    uther
    -0.83
    uph
    -0.82
    ̶
    -0.77
    vier
    -0.77
    ievers
    -0.76
    =-=-=-=-
    -0.75
     Possibly
    -0.75
    --------------------
    -0.73
    utenant
    -0.72
    POSITIVE LOGITS
     namesake
    1.19
     counterparts
    1.17
     inability
    1.15
     reputation
    1.14
     fortunes
    1.14
     cousin
    1.12
     predecessors
    1.10
     antics
    1.09
     penchant
    1.09
     predecessor
    1.07
    Activations Density 0.280%

    No Known Activations