© 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. 4095
    Prev
    Next
    INDEX
    Explanations

    phrases related to responsibility or blame

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate existence or location, particularly focusing on the verb "lie" in various contexts

    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
     minded
    -0.67
    haw
    -0.65
    picking
    -0.60
    aga
    -0.60
     Barney
    -0.58
    panic
    -0.57
    onal
    -0.57
    agers
    -0.56
    oons
    -0.56
    theless
    -0.56
    POSITIVE LOGITS
     dormant
    1.16
     squarely
    1.10
     somewhere
    0.97
     beneath
    0.95
     unexpl
    0.95
    behind
    0.89
     untouched
    0.89
    hidden
    0.85
     buried
    0.84
     unused
    0.84
    Activations Density 0.067%

    No Known Activations