© 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. 23739
    Prev
    Next
    INDEX
    Explanations

    business or marketing terms related to buzz and hype

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "buzz" and its variations, indicating discussions around trending topics or popular content

    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
    ned
    -0.70
     restitution
    -0.68
     Adin
    -0.65
    izations
    -0.65
     Sanct
    -0.65
    isations
    -0.64
     incompatible
    -0.64
    istically
    -0.63
     diam
    -0.60
    INAL
    -0.59
    POSITIVE LOGITS
    saw
    1.53
    words
    1.47
    word
    1.42
    feed
    1.32
    ards
    1.23
    kill
    1.19
    worthy
    1.15
    phrase
    1.13
    ard
    1.05
    arded
    1.03
    Activations Density 0.054%

    No Known Activations