© 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. 23956
    Prev
    Next
    INDEX
    Explanations

    instances of events being described as happening for the first time

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "first" and related phrases indicating precedence or novelty

    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
    rils
    -0.91
    redients
    -0.88
    ooks
    -0.83
    nels
    -0.81
    anders
    -0.80
    adies
    -0.79
    runs
    -0.77
    apters
    -0.77
    acers
    -0.76
    inders
    -0.76
    POSITIVE LOGITS
     time
    1.45
     occasion
    1.19
     foray
    1.10
     instance
    1.10
     indication
    1.10
     occurrence
    0.98
     attempt
    0.96
    time
    0.94
     times
    0.93
     installment
    0.90
    Activations Density 0.078%

    No Known Activations