INDEX

Explanations

phrases expressing surprise or disbelief related to experiences

oai_token-act-pair · gpt-4o-mini

New Auto-Interp

Configuration

fnlp/Llama-Scope-R1-Distill/800M-Slimpajama-0-OpenR1-Math-220k/L15R

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Hzfinfdu/SlimPajama-3B

Features

32,768

Data Type

float32

Hook Name

blocks.15.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

WidgetItem

-0.14

utherland

-0.14

annt

-0.13

.levels

-0.13

an

-0.13

ansen

-0.13

ByUrl

-0.13

 pity

-0.13

 transparency

-0.13

 COUR

-0.12

POSITIVE LOGITS

 surprisingly

0.23

 surprising

0.15

hread

0.15

Ø¯Ùĩ

0.15

_epi

0.15

ostat

0.15

ullo

0.14

ye

0.14

 diffs

0.14

åĿĬ

0.14

Activations Density 0.026%

phrases expressing surprise or disbelief related to experiences

No Comments

No Known Activations