© 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. 11675
    Prev
    Next
    INDEX
    Explanations

    phrases related to setup or action

    oai_token-act-pair · gpt-3.5-turbo

    the phrase "up" 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
    ĸļ
    -0.82
     deeply
    -0.60
     regrets
    -0.57
     harsh
    -0.57
     literal
    -0.57
    çİĭ
    -0.56
    ãģ®å®
    -0.56
     grate
    -0.56
     caution
    -0.55
     sol
    -0.55
    POSITIVE LOGITS
    up
    3.84
    ups
    2.94
    UP
    2.52
    Up
    2.15
    down
    1.79
    upp
    1.62
     Up
    1.60
    upt
    1.53
    upe
    1.48
     ups
    1.41
    Activations Density 0.018%

    No Known Activations