INDEX

Explanations

expressions of gratitude or thankfulness

oai_token-act-pair · gpt-3.5-turbo

expressions of gratitude and thankfulness

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 mathemat

-0.73

 livest

-0.65

Downloadha

-0.64

chio

-0.63

 assum

-0.63

 blat

-0.59

uria

-0.59

aeda

-0.59

astered

-0.58

inates

-0.58

POSITIVE LOGITS

for

0.85

 that

0.81

God

0.80

 they

0.77

giving

0.73

she

0.69

 THEY

0.68

NESS

0.67

FUL

0.67

we

0.66

Activations Density 0.067%

expressions of gratitude or thankfulness

expressions of gratitude and thankfulness

No Comments

No Known Activations