© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 11-RES-JB
    11. 1845
    Prev
    Next
    INDEX
    Explanations

    pronouns or nouns followed by verbs indicating commentary or judgment

    oai_token-act-pair · gpt-3.5-turbo

    references to actors or subjects in a context of criticism or discussion

    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
    hang
    -0.75
    ancial
    -0.70
    ono
    -0.64
    ovo
    -0.63
     Screw
    -0.62
    bleacher
    -0.60
    atible
    -0.58
    rossover
    -0.57
    unity
    -0.57
    ammy
    -0.57
    POSITIVE LOGITS
     termed
    1.53
     called
    1.18
     dubbed
    1.17
     deems
    1.15
     characterized
    1.15
     calls
    1.13
     described
    1.13
     deemed
    1.13
     perceive
    1.12
     perce
    1.12
    Activations Density 0.116%

    No Known Activations