Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 8-RES-JB
    6. 3430
    Prev
    Next
    INDEX
    Explanations

    verbs related to observation or physical action, especially when involving clothing or appearance

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating military presence and condition

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    •interesting
    be/been/was
    •contextual features
    fires on e.g., be/is/was <at a place>. Fires on single token " here"

    No Comments

    Negative Logits
     Moroc
    -0.79
     Vaugh
    -0.73
     Ambro
    -0.71
     Sutherland
    -0.69
     Azerb
    -0.67
     Frie
    -0.64
     Seym
    -0.63
     Vanderbilt
    -0.61
    Compar
    -0.60
    ãĥ¼ãĥĨ
    -0.60
    POSITIVE LOGITS
    "?
    0.70
    tics
    0.69
    !!!!
    0.68
    '?
    0.67
    !?
    0.67
     !
    0.67
     tho
    0.65
     ye
    0.64
    !
    0.64
    !!!
    0.63
    Activations Density 0.581%

    No Known Activations