INDEX

Explanations

explanations or reasons

oai_token-act-pair · gpt-3.5-turbo

references to the concept of explanations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ymph

-0.87

inal

-0.80

illet

-0.80

yss

-0.76

ibaba

-0.75

estial

-0.75

semble

-0.73

oned

-0.70

vals

-0.69

ony

-0.69

POSITIVE LOGITS

WHY

1.05

 explanation

0.99

 explanations

0.98

why

0.97

ãĤ©

0.86

 explan

0.80

 explaining

0.80

Origin

0.80

 answ

0.79

why

0.78

Activations Density 0.024%

explanations or reasons

references to the concept of explanations

No Comments

No Known Activations