INDEX

Explanations

dates and time-related information

oai_token-act-pair · gpt-3.5-turbo

temporal references or indicators related to time and duration

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Tray

-0.52

mosp

-0.44

WAR

-0.44

nar

-0.42

 Roose

-0.41

bre

-0.41

Ble

-0.40

 TOUR

-0.40

Var

-0.40

 Sync

-0.40

POSITIVE LOGITS

.''.

0.70

 onwards

0.67

ãĢĤ

0.65

.'

0.59

é¾įå¥ĳå£«

0.59

 guiActiveUnfocused

0.59

.[

0.57

.''

0.56

'.

0.55

Activations Density 0.941%

dates and time-related information

temporal references or indicators related to time and duration

No Comments

No Known Activations