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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 827
    Prev
    Next
    INDEX
    Explanations

    phrases related to biological processes and cellular components

    oai_token-act-pair · gpt-3.5-turbo

    instances of the verb "to be" in various forms, indicating existence or presence

    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
     Dollars
    -0.81
     IMAGES
    -0.75
     Prices
    -0.75
     Purchase
    -0.70
     Nights
    -0.70
     Lights
    -0.69
    erguson
    -0.67
     Wish
    -0.67
     Fare
    -0.67
     Hours
    -0.66
    POSITIVE LOGITS
     implicated
    1.18
     inhibited
    1.12
     responsible
    1.12
     oxid
    1.11
     encoded
    1.10
     chemically
    1.10
     synthes
    1.09
     mutated
    1.09
     metabol
    1.06
     capable
    1.06
    Activations Density 0.110%

    No Known Activations