© 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. 7854
    Prev
    Next
    INDEX
    Explanations

    people's names along with their associated titles or roles

    oai_token-act-pair · gpt-3.5-turbo

    instances of the verb "was" indicating past experiences or states of being

    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
    eva
    -0.78
    olutions
    -0.78
     ado
    -0.70
    itage
    -0.70
    ieth
    -0.67
    ]]
    -0.66
     Needs
    -0.64
    idium
    -0.63
    haven
    -0.63
     Retrieved
    -0.63
    POSITIVE LOGITS
     nicknamed
    1.11
     fascinated
    1.10
     obsessed
    1.10
     able
    1.10
     known
    1.03
     proficient
    1.02
     bullied
    1.01
     adept
    1.01
     rumored
    0.99
     fluent
    0.98
    Activations Density 0.272%

    No Known Activations