© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 22586
    Prev
    Next
    INDEX
    Explanations

    phrases indicating a group or collection of people or entities

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate composition or group affiliation

    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
    bringer
    -0.79
    Url
    -0.74
    aura
    -0.66
    Weapon
    -0.64
    bra
    -0.64
    士
    -0.63
    imation
    -0.62
    TV
    -0.62
    catentry
    -0.62
    escape
    -0.61
    POSITIVE LOGITS
     representatives
    1.15
     members
    1.03
     dozens
    0.95
     partners
    0.94
     contributors
    0.94
     collaborators
    0.94
     stakeholders
    0.92
     five
    0.88
     participants
    0.87
    members
    0.87
    Activations Density 0.440%

    No Known Activations