© 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. 10-RES-JB
    11. 9887
    Prev
    Next
    INDEX
    Explanations

    text related to expressing support for various causes or individuals

    oai_token-act-pair · gpt-3.5-turbo

    references to political support and alliances

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ipl
    -0.75
     pandemonium
    -0.70
    wow
    -0.63
     prest
    -0.61
     seams
    -0.59
     overlap
    -0.58
    abulary
    -0.57
     mirrors
    -0.57
    dust
    -0.56
     Affect
    -0.55
    POSITIVE LOGITS
     boycott
    0.98
     initiatives
    0.95
     efforts
    0.91
     legalization
    0.90
     candidacy
    0.90
     financially
    0.89
     uncond
    0.89
     underdog
    0.87
     against
    0.85
     advoc
    0.84
    Activations Density 0.376%

    No Known Activations