© 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. 3572
    Prev
    Next
    INDEX
    Explanations

    phrases related to trying to reach someone or obtain information without success

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "could" indicating possibility or inability

    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
     Ages
    -0.66
     Gran
    -0.63
     Outer
    -0.61
     Cortex
    -0.60
     Dunham
    -0.60
     Overs
    -0.59
     Dek
    -0.58
    ļé
    -0.58
    ItemImage
    -0.57
     Learns
    -0.57
    POSITIVE LOGITS
    nt
    1.18
     scarcely
    1.14
     not
    1.13
     hardly
    1.13
     afford
    1.12
    NOT
    1.04
     feas
    1.01
    not
    0.98
     muster
    0.95
     barely
    0.93
    Activations Density 0.074%

    No Known Activations