INDEX

Explanations

punctuation marks that indicate pauses or breaks in text

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

Tokens preceding conjunctions or adverbs

np_acts-logits-general · gemini-2.0-flash

conjunctions especially after hyphens

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

 honom

-0.55

(?)

-0.55

(%)

-0.53

principalTable

-0.52

stb

-0.48

qai

-0.47

osv

-0.46

(?)

-0.46

[?]

-0.45

Them

-0.43

POSITIVE LOGITS

 hence

1.36

 although

1.31

 namely

1.25

hence

1.22

 they

1.21

although

1.15

namely

1.14

 albeit

1.11

 there

1.10

 indeed

1.06

Activations Density 0.246%

punctuation marks that indicate pauses or breaks in text

Tokens preceding conjunctions or adverbs

conjunctions especially after hyphens

No Comments

No Known Activations