© 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. 1-RES-JB
    11. 24252
    Prev
    Next
    INDEX
    Explanations

    words and phrases related to being rewarded for certain actions or behaviors

    oai_token-act-pair · gpt-3.5-turbo

    terms related to rewarding and punishing actions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    frame
    -0.75
    space
    -0.72
    sie
    -0.72
    frames
    -0.68
    cell
    -0.66
    CON
    -0.66
    orig
    -0.65
    alter
    -0.65
    aug
    -0.65
    issues
    -0.65
    POSITIVE LOGITS
     rewarded
    1.54
     rewarding
    1.15
     rewards
    1.11
     reward
    1.08
     tremend
    0.98
    nesday
    0.98
     incentiv
    0.89
     veter
    0.87
     reap
    0.87
     showc
    0.87
    Activations Density 0.015%

    No Known Activations