© 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. 12-RES-JB
    11. 4398
    Prev
    Next
    INDEX
    Explanations

    phrases related to acts of ignoring or disregarding something

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

    instances of the word "Ign" or variations related to the concept of ignoring

    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
     Tale
    -0.64
     Doodle
    -0.63
    DAQ
    -0.62
     Morph
    -0.60
     nationals
    -0.60
     Quinn
    -0.59
    TON
    -0.59
     liabilities
    -0.59
     feelings
    -0.58
     Catal
    -0.57
    POSITIVE LOGITS
    acio
    1.46
    oring
    1.25
    omin
    1.23
    ition
    1.21
    oreAnd
    1.14
    atz
    1.06
    orable
    1.05
    ored
    1.05
    ores
    1.04
    orem
    1.02
    Activations Density 0.056%

    No Known Activations