© 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 · Initial Gemma-2B-IT SAEs
    5. Gemma-2B-IT
    7. Residual Stream
    9. 12-RES-JB
    11. 7721
    Prev
    Next
    INDEX
    Explanations

    verbs related to action and behaviors in instructions or explanations, such as "simply take," "drop it," "have it work," "go out on a limb," and "make a diagnosis."

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    jbloom/Gemma-2b-IT-Residual-Stream-SAEs/gemma_2b_it_blocks.12.hook_resid_post_16384
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    HuggingFaceFW/fineweb
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.12.hook_resid_post
    Hook Layer
    12
    Architecture
    standard
    Context Size
    1,024
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    1253
    +0.11
    0.3%
    1978
    +0.10
    0.3%
    845
    +0.08
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1261
    +0.11
    0.02
    911
    +0.10
    0.03
    1941
    +0.08
    0.02
    Negative Logits
     shenan
    -1.30
     milf
    -1.26
     apprehen
    -1.23
     unspeak
    -1.14
     maneu
    -1.13
     depic
    -1.12
     reluct
    -1.12
     Intere
    -1.08
     cuck
    -1.07
     snoopy
    -1.05
    POSITIVE LOGITS
    <bos>
    0.79
     ProtoMessage
    0.67
     soggior
    0.64
     незавершена
    0.62
     مرئيه
    0.59
    Personendaten
    0.59
     nor
    0.58
     anymore
    0.58
     grans
    0.57
     AssemblyVersion
    0.56
    Activations Density 0.615%

    No Known Activations