© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 692
    Prev
    Next
    INDEX
    Explanations

    specific phrases indicating hypothetical situations and emotions

    oai_token-act-pair · gpt-3.5-turbo

    instances of the verb "be" in various forms and contexts

    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
     converge
    -0.68
     procedure
    -0.67
     originate
    -0.64
    aceutical
    -0.64
     residue
    -0.63
     scrimmage
    -0.63
     spectrum
    -0.61
    osate
    -0.60
     disperse
    -0.60
     arise
    -0.59
    POSITIVE LOGITS
     able
    1.15
    friends
    1.07
     thankful
    0.93
    getting
    0.93
     glad
    0.87
    acons
    0.87
     happiest
    0.87
    heading
    0.87
    league
    0.85
    orc
    0.85
    Activations Density 0.235%

    No Known Activations