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

    Neuronpedia

    Jacobian LensNEW
    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. 0-RES-JB
    11. 22161
    Prev
    Next
    INDEX
    Explanations

    links or instructions indicated by the word 'following'

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "following."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    utm
    -0.63
    ello
    -0.60
    ebook
    -0.60
    vice
    -0.60
    inski
    -0.58
    arta
    -0.58
     nowhere
    -0.57
    Frameworks
    -0.56
    clerosis
    -0.56
    imony
    -0.56
    POSITIVE LOGITS
     following
    3.62
    follow
    1.99
     Following
    1.90
     preceding
    1.57
     follows
    1.57
     foregoing
    1.45
    Following
    1.41
     succeeding
    1.35
     follow
    1.25
     resulting
    1.19
    Activations Density 0.031%

    No Known Activations