© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 9-RES-JB
    6. 13123
    Prev
    Next
    INDEX
    Explanations

    pronouns and actions in a sequence of events involving deception, conflict, and law enforcement

    oai_token-act-pair · gpt-3.5-turbo

    references to characters involved in a conflict or negotiation

    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
     Consortium
    -0.75
    20439
    -0.73
     Prelude
    -0.71
    undrum
    -0.66
     Initiative
    -0.66
    igmatic
    -0.65
    eaturing
    -0.64
     Summit
    -0.63
     Uncharted
    -0.63
    pires
    -0.62
    POSITIVE LOGITS
     knew
    1.18
     gave
    1.18
     swore
    1.15
     apologized
    1.15
     didn
    1.12
     forgot
    1.11
     pretended
    1.10
     did
    1.09
     threw
    1.09
     waited
    1.08
    Activations Density 0.435%

    No Known Activations