INDEX

Explanations

phrases introducing a new topic or providing additional information

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate hypothetical scenarios or conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

elta

-0.90

pez

-0.81

¥ŀ

-0.79

assi

-0.77

erer

-0.76

resent

-0.76

cribed

-0.74

ivated

-0.73

kil

-0.73

ieved

-0.70

POSITIVE LOGITS

you

0.75

 mater

0.71

DragonMagazine

0.69

fy

0.67

 they

0.67

 circumstances

0.67

 anyone

0.65

we

0.65

 cases

0.64

imoto

0.62

Activations Density 0.022%

phrases introducing a new topic or providing additional information

phrases that indicate hypothetical scenarios or conditions

No Comments

No Known Activations