© 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. 11836
    Prev
    Next
    INDEX
    Explanations

    German proper nouns

    oai_token-act-pair · gpt-3.5-turbo

    the presence of the term "Der" and its variations, indicating a focus on German sources or references

    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
     [|
    -0.69
    hearted
    -0.60
     overdue
    -0.59
     peanuts
    -0.58
     aimed
    -0.57
     commissions
    -0.56
     fixtures
    -0.56
     coincidence
    -0.56
     instituted
    -0.56
     confidence
    -0.56
    POSITIVE LOGITS
    bys
    1.17
    isively
    1.13
    ivation
    1.11
     Spiegel
    1.10
    ricks
    1.07
    rick
    1.02
    anged
    1.02
    ived
    1.01
    iving
    1.00
    isive
    1.00
    Activations Density 0.043%

    No Known Activations