INDEX

Explanations

elements related to significant narrative shifts or changes in context

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

Code, math, or punctuation

np_acts-logits-general · gemini-2.0-flash

animation

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

 ujednoznacz

-0.75

 المعيارى

-0.64

 propOrder

-0.59

 keynote

-0.56

 الحره

-0.55

évaluateur

-0.54

PYX

-0.53

 watershed

-0.51

ckeye

-0.50

wness

-0.50

POSITIVE LOGITS

 animation

0.76

 animations

0.71

 anim

0.69

 Animation

0.67

 player

0.66

 enemy

0.64

animation

0.61

 Player

0.60

 Animations

0.59

 sprite

0.58

Activations Density 0.349%

elements related to significant narrative shifts or changes in context

Code, math, or punctuation

animation

No Comments

No Known Activations