© 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. 10-RES-JB
    11. 16175
    Prev
    Next
    INDEX
    Explanations

    phrases related to superlatives or comparisons in terms of speed or rank

    oai_token-act-pair · gpt-3.5-turbo

    superlative adjectives and phrases indicating rank or growth

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     mediation
    -0.67
     disguise
    -0.67
     simplicity
    -0.67
     haze
    -0.66
     indifference
    -0.66
     Daylight
    -0.65
     Marble
    -0.64
     flashback
    -0.64
     glasses
    -0.63
     assurances
    -0.62
    POSITIVE LOGITS
    ranked
    1.50
    selling
    1.39
    ever
    1.38
    performing
    1.38
    known
    1.35
    rated
    1.33
    gross
    1.30
    serving
    1.27
    funded
    1.24
    paid
    1.23
    Activations Density 0.064%

    No Known Activations