INDEX

Explanations

repetitive phrases emphasizing the word "all"

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

"all" followed by a variety of words

np_acts-logits-general · gemini-2.0-flash

all sorts of

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

 leaſt

-0.69

AsUp

-0.69

Efq

-0.66

CDCl

-0.66

 Perſ

-0.64

omiast

-0.64

 angekommen

-0.61

 Others

-0.61

rmtree

-0.61

 pathlib

-0.61

POSITIVE LOGITS

 sorts

0.96

 kinds

0.79

kinds

0.65

 sortes

0.65

 manner

0.64

 anyone

0.51

Kinds

0.51

 Kinds

0.50

we

0.46

 that

0.45

Activations Density 0.065%

repetitive phrases emphasizing the word "all"

"all" followed by a variety of words

all sorts of

No Comments

No Known Activations