© 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. 16872
    Prev
    Next
    INDEX
    Explanations

    instances of events happening repeatedly or multiple times

    oai_token-act-pair · gpt-3.5-turbo

    instances related to multiple occurrences of events or situations

    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
    MIT
    -0.63
     Translation
    -0.62
    rig
    -0.59
    REL
    -0.58
    oiler
    -0.55
    pan
    -0.55
    belt
    -0.53
    killer
    -0.52
    hed
    -0.52
    cock
    -0.52
    POSITIVE LOGITS
     occasions
    3.92
     occasion
    2.62
     instances
    1.51
     occas
    1.35
     situations
    1.29
     times
    1.28
     evenings
    1.28
     occurrences
    1.26
     outings
    1.26
     gatherings
    1.17
    Activations Density 0.005%

    No Known Activations