© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 12-RES-JB
    11. 2669
    Prev
    Next
    INDEX
    Explanations

    political statements and discussions, particularly critical views related to government decisions and actions

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot

    forms of the verb 'to be' in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 12-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_post
    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_post
    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
    icipated
    -0.86
    »Ĵ
    -0.85
    avez
    -0.79
    =-=-=-=-
    -0.78
    requent
    -0.78
    jandro
    -0.78
    acters
    -0.77
    hess
    -0.77
    pleted
    -0.74
    urry
    -0.72
    POSITIVE LOGITS
     complicit
    1.23
     hypoc
    1.22
     waging
    1.15
     incompetent
    1.12
     conspiring
    1.12
     rigging
    1.11
     clueless
    1.09
     ignoring
    1.08
     incapable
    1.07
     ruining
    1.05
    Activations Density 0.265%

    No Known Activations