© 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. 4-RES-JB
    11. 6096
    Prev
    Next
    INDEX
    Explanations

    references to actions or events happening behind closed doors

    oai_token-act-pair · gpt-3.5-turbo

    mentions of "doors" in various contexts related to access or closure

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

    No Comments

    Negative Logits
    TY
    -0.78
    olson
    -0.74
    TING
    -0.70
     Astro
    -0.67
     Coastal
    -0.66
    lihood
    -0.66
    TX
    -0.64
    ollah
    -0.63
     Hots
    -0.63
    CTV
    -0.63
    POSITIVE LOGITS
     doors
    1.12
    doors
    1.10
    holes
    0.99
    door
    0.96
    bell
    0.93
    hole
    0.88
    pring
    0.87
    steps
    0.84
    opening
    0.82
    mith
    0.81
    Activations Density 0.011%

    No Known Activations