INDEX

Explanations

words related to emotional reactions like fascination, excitement, and anger

oai_token-act-pair · gpt-3.5-turbo

emotional reactions and responses related to engagement and intrigue

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

©¶æ

-0.81

porary

-0.69

itled

-0.66

 forearm

-0.65

uts

-0.65

glas

-0.63

 slips

-0.63

nect

-0.62

uben

-0.62

abs

-0.61

POSITIVE LOGITS

ingly

1.05

 onlook

0.95

 audiences

0.92

us

0.89

me

0.85

him

0.78

 passers

0.78

 viewers

0.77

 readers

0.74

 sensibilities

0.74

Activations Density 0.182%

words related to emotional reactions like fascination, excitement, and anger

emotional reactions and responses related to engagement and intrigue

No Comments

No Known Activations