© 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. 5382
    Prev
    Next
    INDEX
    Explanations

    proper nouns of people and places

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "come" and its variations related to arrival or origin

    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
     Delete
    -0.63
     Individual
    -0.57
     Accuracy
    -0.56
     Repeat
    -0.55
     mitigation
    -0.55
     Defense
    -0.54
     Balanced
    -0.53
    IFT
    -0.53
     Genie
    -0.53
    Defense
    -0.52
    POSITIVE LOGITS
    os
    0.80
     ashore
    0.77
    iments
    0.76
    iment
    0.76
     forth
    0.76
    ographically
    0.75
     hither
    0.74
    ortal
    0.73
    olog
    0.73
    ographed
    0.71
    Activations Density 0.066%

    No Known Activations