INDEX

Explanations

sections that include legal language and procedural references

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.13.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Roskov

-0.82

tvguidetime

-0.79

StoryboardSegue

-0.76

ynb

-0.74

featureID

-0.71

ArrowToggle

-0.69

webElementXpaths

-0.69

ceased

-0.67

 lenker

-0.67

reportWebVitals

-0.66

POSITIVE LOGITS

 than

0.77

Another

0.60

Again

0.60

 kuin

0.56

 lagi

0.55

again

0.55

 nữa

0.54

 again

0.54

than

0.53

 Another

0.53

Activations Density 0.154%

sections that include legal language and procedural references

No Comments

No Known Activations