INDEX

Explanations

the word "into" in various contexts

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

into a

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

 enfans

-0.98

 purpoſe

-0.77

 Juifs

-0.75

 Grecs

-0.75

 avoient

-0.75

SEGUIR

-0.73

 étoient

-0.71

 ſta

-0.71

 juges

-0.70

 ainfi

-0.70

POSITIVE LOGITS

 into

1.28

 INTO

1.15

 Into

1.08

into

1.07

Into

1.05

INTO

0.81

 onto

0.72

 dans

0.61

 inside

0.58

isIn

0.57

Activations Density 0.060%

the word "into" in various contexts

into a

No Comments

No Known Activations

the word "into" in various contexts

into a

No Comments

No Known Activations