© 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 EvalsExportsAPI 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. 3-RES-JB
    11. 21008
    Prev
    Next
    INDEX
    Explanations

    phrases with the structure "XY-face," where X and Y are typically prepositions or possessive pronouns

    oai_token-act-pair · gpt-3.5-turbo

    negative phrases related to online and in-person experiences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Dickinson
    -0.81
     Norris
    -0.79
     Compass
    -0.74
    ulhu
    -0.73
     Confederation
    -0.70
    ãģ®å®
    -0.69
     Tempest
    -0.68
     Rivals
    -0.67
     Burr
    -0.67
     AVG
    -0.65
    POSITIVE LOGITS
    sized
    1.27
    based
    1.23
    centered
    1.10
    level
    1.07
    shaped
    1.05
    related
    1.04
    minded
    1.03
    focused
    1.03
    to
    1.02
    scale
    1.02
    Activations Density 0.333%

    No Known Activations