© 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. 16558
    Prev
    Next
    INDEX
    Explanations

    phrases related to activities done continuously over a period of time

    oai_token-act-pair · gpt-3.5-turbo

    the word "the" and phrases that include repeated use of "the" along with particular contexts

    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
    oma
    -0.81
    CTV
    -0.71
    interstitial
    -0.67
    TextColor
    -0.65
    BIL
    -0.64
     Franch
    -0.63
    nit
    -0.63
    acons
    -0.62
    NAME
    -0.62
    Chair
    -0.61
    POSITIVE LOGITS
     day
    1.40
     week
    1.23
     year
    1.20
     evening
    1.20
     duration
    1.19
     semester
    1.18
     afternoon
    1.15
     process
    1.14
     course
    1.12
     night
    1.11
    Activations Density 0.119%

    No Known Activations