© 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. 5618
    Prev
    Next
    INDEX
    Explanations

    political events and conflicts involving various countries

    oai_token-act-pair · gpt-3.5-turbo

    statements regarding accusations and responsibility in conflict situations

    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
     partName
    -0.75
     Anxiety
    -0.72
     Mania
    -0.69
     Trance
    -0.69
     Remastered
    -0.68
    natureconservancy
    -0.68
     ILCS
    -0.68
    Disney
    -0.67
    rified
    -0.67
     FANTASY
    -0.66
    POSITIVE LOGITS
     massacres
    1.33
     atrocities
    1.33
     killings
    1.27
     bombings
    1.16
     massac
    1.14
     assass
    1.13
     genocide
    1.12
     killing
    1.11
     abduct
    1.11
     abuses
    1.07
    Activations Density 0.255%

    No Known Activations