© 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. 16611
    Prev
    Next
    INDEX
    Explanations

    references prompting the reader to stay updated or in touch for future information

    oai_token-act-pair · gpt-3.5-turbo

    phrases urging the audience to remain updated or connected

    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
    £ı
    -0.72
    OB
    -0.71
     unilaterally
    -0.68
    Ĭ±
    -0.66
     disadvant
    -0.66
    ocr
    -0.66
    isson
    -0.66
    alogy
    -0.65
    naires
    -0.62
    Failure
    -0.61
    POSITIVE LOGITS
     tuned
    1.50
     updated
    1.00
     informed
    0.92
     Tun
    0.88
     tuning
    0.88
    informed
    0.83
     vigilant
    0.81
     subscribed
    0.80
     classy
    0.79
     logged
    0.77
    Activations Density 0.077%

    No Known Activations