INDEX

Explanations

words and phrases expressing surprise

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

yakarta

-0.59

toJSONString

-0.56

ędzy

-0.52

TacToe

-0.51

styrelsen

-0.50

setViewportView

-0.48

erdings

-0.48

 benzina

-0.48

brengen

-0.47

erobe

-0.47

POSITIVE LOGITS

 surprise

2.25

 Surprise

2.05

 surprises

2.03

 surprised

2.00

Surprise

1.92

surprise

1.89

 surprising

1.72

 overras

1.71

surprised

1.70

 Überras

1.69

Activations Density 0.151%

words and phrases expressing surprise

No Comments

No Known Activations