© 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. 23974
    Prev
    Next
    INDEX
    Explanations

    phrases related to premium services and upgrades

    oai_token-act-pair · gpt-3.5-turbo

    promotional phrases related to subscriptions and upgrades

    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
     Cors
    -0.65
     Cobb
    -0.63
     Fors
    -0.61
     Corm
    -0.61
    lict
    -0.61
    iosyncr
    -0.60
    unts
    -0.60
     downed
    -0.58
    vg
    -0.58
     Lyme
    -0.57
    POSITIVE LOGITS
    prem
    0.75
     ads
    0.70
    jew
    0.68
    usalem
    0.66
    ppo
    0.65
    Content
    0.63
    audio
    0.62
    thood
    0.60
     advertising
    0.60
    izons
    0.59
    Activations Density 0.022%

    No Known Activations