© 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. 3-RES-JB
    11. 7303
    Prev
    Next
    INDEX
    Explanations

    verbs that imply a progression or reaching a peak

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate the conclusion or climax of events

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Neighbor
    -0.83
    nery
    -0.78
    eling
    -0.71
    aky
    -0.71
    alcohol
    -0.68
    rian
    -0.68
    agg
    -0.68
    uns
    -0.67
    isen
    -0.66
    asel
    -0.66
    POSITIVE LOGITS
     culminating
    1.11
     culminated
    0.92
     culmin
    0.89
    ãĤ¤ãĥĪ
    0.81
     culmination
    0.79
    CRIP
    0.77
     pandemonium
    0.76
     climax
    0.74
    ADRA
    0.74
    ODE
    0.73
    Activations Density 0.011%

    No Known Activations