INDEX

Explanations

variations of the word "whether" indicating uncertainty or choice

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Cooley

-0.73

 Millard

-0.69

ldc

-0.64

 Pires

-0.62

actus

-0.62

ioe

-0.61

 Reid

-0.61

珑

-0.60

 Hayden

-0.60

 Feind

-0.60

POSITIVE LOGITS

 WHETHER

1.93

whether

1.82

 whether

1.78

 Whether

1.76

Whether

1.74

 apakah

1.11

WHE

1.08

 jestli

1.01

enderror

1.01

whe

0.97

Activations Density 0.090%

variations of the word "whether" indicating uncertainty or choice

No Comments

No Known Activations