© 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. 9-RES-JB
    11. 9061
    Prev
    Next
    INDEX
    Explanations

    entities followed by verbs in the past tense

    oai_token-act-pair · gpt-3.5-turbo

    references to authorities or verification processes involving vendors or individuals

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    9
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     clich
    -0.80
     vitality
    -0.77
     Highlights
    -0.73
     highs
    -0.72
     laughter
    -0.71
     fascinating
    -0.70
    lust
    -0.70
     immortality
    -0.68
     simplicity
    -0.68
    zees
    -0.68
    POSITIVE LOGITS
     oversee
    1.28
     notified
    1.19
     superv
    1.18
     intervened
    1.15
     contacted
    1.14
     oversees
    1.13
     oversaw
    1.10
     administer
    1.08
     jurisdiction
    1.06
     consulted
    1.06
    Activations Density 0.704%

    No Known Activations