© 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. 0-RES-JB
    11. 21507
    Prev
    Next
    INDEX
    Explanations

    continuous prompts to read more of a main story

    oai_token-act-pair · gpt-3.5-turbo

    instances of the phrase "Continue reading" in various contexts

    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
     corpor
    -0.69
     pert
    -0.68
    ball
    -0.65
     gig
    -0.61
     quad
    -0.60
     mobile
    -0.60
     sources
    -0.60
     knowledge
    -0.59
     ratios
    -0.59
     drunk
    -0.58
    POSITIVE LOGITS
     Continue
    3.89
    Continue
    2.24
    continue
    1.75
     CONTIN
    1.62
     Continued
    1.45
     Proceed
    1.33
     continue
    1.27
     Continuing
    1.27
     Finish
    1.18
     Begin
    1.14
    Activations Density 0.020%

    No Known Activations