© 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 EvalsExportsinterp-engineNEWAPI 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. 11-RES-JB
    11. 11622
    Prev
    Next
    INDEX
    Explanations

    phrases related to quantity or intensity expressed through the use of the word 'bit' and its variations

    oai_token-act-pair · gpt-3.5-turbo

    phrases that refer to small amounts or moderate quantities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ents
    -1.00
    uers
    -0.99
    ourses
    -0.96
    brates
    -0.94
    ards
    -0.93
    estones
    -0.93
    ires
    -0.92
    ongs
    -0.92
    Cs
    -0.91
    ales
    -0.91
    POSITIVE LOGITS
     luck
    1.06
     flair
    1.00
     realism
    1.00
     irony
    0.99
     backstory
    0.94
     sweetness
    0.93
     mischief
    0.93
     humour
    0.92
     elbow
    0.90
     spice
    0.89
    Activations Density 0.077%

    No Known Activations