© 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. 13169
    Prev
    Next
    INDEX
    Explanations

    irregular verb forms ending in 'st' or 't' preceded by 'for' such as "forgot" or "left"

    oai_token-act-pair · gpt-3.5-turbo

    words associated with action or command verbs

    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
    bour
    -0.71
     differentiated
    -0.66
     pitched
    -0.58
    Fri
    -0.58
    Deal
    -0.57
    rooms
    -0.56
    rose
    -0.56
    ibu
    -0.56
    Higher
    -0.55
     pee
    -0.55
    POSITIVE LOGITS
     sake
    1.05
    iencies
    0.94
     reasons
    0.90
    iture
    0.84
    ummies
    0.82
    eworks
    0.78
    iveness
    0.75
    ulum
    0.75
    pection
    0.74
     purposes
    0.73
    Activations Density 0.051%

    No Known Activations