© 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. 10-RES-JB
    11. 20413
    Prev
    Next
    INDEX
    Explanations

    phrases related to competitions or sports events with a specific format such as "best of seven" or "best of three"

    oai_token-act-pair · gpt-3.5-turbo

    references to best-of series in competitions or rankings

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Helic
    -0.70
     Arabian
    -0.68
     Carib
    -0.67
     Creator
    -0.64
     Chairman
    -0.63
     horizon
    -0.63
     Aval
    -0.63
     trough
    -0.63
     Indigo
    -0.63
     brim
    -0.62
    POSITIVE LOGITS
    selling
    1.28
    sounding
    1.26
    looking
    1.25
    known
    1.19
    sized
    1.18
    nat
    1.13
    luck
    1.12
    performing
    1.12
    enough
    1.10
    paying
    1.08
    Activations Density 0.029%

    No Known Activations