INDEX

Explanations

words expressing uncertainty or speculation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

喰

-0.47

Rout

-0.45

 subordination

-0.45

rout

-0.44

geführten

-0.43

 Rout

-0.43

 OrderedDict

-0.43

orithmic

-0.42

 oorlog

-0.42

 operations

-0.42

POSITIVE LOGITS

perhaps

0.95

maybe

0.93

Maybe

0.93

 Maybe

0.91

 maybe

0.90

 Perhaps

0.88

 perhaps

0.86

Perhaps

0.81

anskje

0.80

 quizá

0.71

Activations Density 0.040%

words expressing uncertainty or speculation

No Comments

No Known Activations