© 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 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. 8-RES-JB
    11. 21692
    Prev
    Next
    INDEX
    Explanations

    adjectives with negative connotations, particularly related to social and ethical issues

    oai_token-act-pair · gpt-3.5-turbo

    instances of strong opinions and assertions about social issues

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    intend
    -0.72
     orbits
    -0.70
    çīĪ
    -0.68
     specs
    -0.68
     Explorer
    -0.68
    phrine
    -0.67
     slots
    -0.65
     specifications
    -0.65
     biod
    -0.65
    akeru
    -0.64
    POSITIVE LOGITS
     perpetrated
    1.36
     rampant
    1.30
     rife
    1.22
     abound
    1.08
     contagious
    1.07
     punishable
    1.02
     inflicted
    0.98
     pervasive
    0.97
     prevalent
    0.92
     endemic
    0.92
    Activations Density 0.402%

    No Known Activations