© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 20425
    Prev
    Next
    INDEX
    Explanations

    words related to rapid movement or distance adjustment, such as "zoom" and "swoop"

    oai_token-act-pair · gpt-3.5-turbo

    terms related to movement or rapid action, particularly in contexts involving zooming or swooping

    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
    alg
    -0.83
    arten
    -0.78
     Chall
    -0.77
    administ
    -0.74
    Marie
    -0.74
     Danger
    -0.74
     Welch
    -0.73
     Charm
    -0.72
     Auth
    -0.72
     Berm
    -0.70
    POSITIVE LOGITS
     zoom
    2.18
     Shia
    1.93
     Zoom
    1.55
     swoop
    1.45
     Shiite
    1.40
     squat
    1.33
     enlarge
    1.29
     sway
    1.25
     zo
    1.15
     magnification
    1.05
    Activations Density 0.036%

    No Known Activations