© 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. 18822
    Prev
    Next
    INDEX
    Explanations

    phrases related to specific items or objects

    oai_token-act-pair · gpt-3.5-turbo

    definite articles or determiners 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
    FILE
    -0.85
    AGE
    -0.76
    ade
    -0.73
    Background
    -0.72
    According
    -0.72
    ees
    -0.72
    Episode
    -0.72
    aran
    -0.72
     beforehand
    -0.71
    Britain
    -0.70
    POSITIVE LOGITS
     occasional
    1.04
     dreaded
    1.00
     ones
    0.96
     slightest
    0.91
     obligatory
    0.88
     aforementioned
    0.86
     downright
    0.85
     endless
    0.82
     smallest
    0.82
     latter
    0.81
    Activations Density 0.238%

    No Known Activations