© 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. 4-RES-JB
    11. 7353
    Prev
    Next
    INDEX
    Explanations

    words related to debates and tensions, often associated with conflict or opposition

    oai_token-act-pair · gpt-3.5-turbo

    various economic and social concepts, particularly relating to distress and conditions impacting individuals or communities

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

    No Comments

    Negative Logits
     Accountability
    -0.69
     Own
    -0.59
     Balanced
    -0.58
     Handbook
    -0.58
     Parish
    -0.58
     Shuttle
    -0.56
     Lines
    -0.56
     Lobby
    -0.56
     Transparency
    -0.56
     Marketplace
    -0.56
    POSITIVE LOGITS
    agra
    0.69
    ect
    0.66
    omorph
    0.61
    id
    0.60
    Äĵ
    0.59
    ont
    0.58
    ole
    0.58
    ceed
    0.58
    oles
    0.57
    ol
    0.57
    Activations Density 0.593%

    No Known Activations