INDEX

Explanations

references to personal experiences and feelings within a contextual narrative

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

"in a" followed by an adjective or adverb

np_acts-logits-general · gemini-2.0-flash

sort of kind

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

出版年

-0.72

 propOrder

-0.70

^(@)

-0.68

 toller

-0.64

 Majefty

-0.64

arakhand

-0.64

 tolerance

-0.62

ſelf

-0.61

 Toler

-0.60

 tolerancia

-0.60

POSITIVE LOGITS

 sorts

1.22

 sort

1.10

sort

0.87

 sorta

0.77

 Sort

0.77

 gewisser

0.75

Kinda

0.71

 essentially

0.70

 kind

0.70

 sortes

0.69

Activations Density 0.171%

references to personal experiences and feelings within a contextual narrative

"in a" followed by an adjective or adverb

sort of kind

No Comments

No Known Activations