© 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. 12-RES-JB
    11. 3308
    Prev
    Next
    INDEX
    Explanations

    political statements and discussions related to current events and policies, especially in the context of community and political action

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot

    collective expressions of concern or demands from a community

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 12-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_post
    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_post
    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
    tains
    -0.80
    typ
    -0.74
    uces
    -0.73
    ulence
    -0.69
    parts
    -0.67
    Magikarp
    -0.67
    rocket
    -0.65
    lights
    -0.64
    Textures
    -0.63
    WithNo
    -0.63
    POSITIVE LOGITS
    akening
    1.29
     owe
    1.16
     need
    1.14
     must
    1.10
     cannot
    1.08
    've
    1.05
     ought
    1.04
    're
    1.04
     oppose
    1.00
    akens
    1.00
    Activations Density 0.200%

    No Known Activations