INDEX

Explanations

phrases and words that indicate questions and inquiry

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

borderSide

-0.60

udessa

-0.57

Опис

-0.54

effort

-0.54

MBOL

-0.50

haps

-0.50

sighted

-0.49

 effort

-0.49

izd

-0.49

Effort

-0.49

POSITIVE LOGITS

 question

2.27

 questions

2.25

 Question

1.93

 Questions

1.92

question

1.92

questions

1.86

Questions

1.85

Question

1.79

 Frage

1.74

 QUESTION

1.70

Activations Density 0.633%

phrases and words that indicate questions and inquiry

No Comments

No Known Activations