INDEX

Explanations

top rankings or lists

oai_token-act-pair · gpt-3.5-turbo

references to ordinal rankings or lists

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

walking

-0.75

 Introduced

-0.74

joy

-0.72

Hik

-0.70

uala

-0.70

igham

-0.69

 Seym

-0.66

uron

-0.63

elcome

-0.61

jab

-0.61

POSITIVE LOGITS

nings

0.87

 contenders

0.87

 earners

0.85

%,

0.81

%:

0.78

%.

0.74

 digits

0.73

 percent

0.70

 rankings

0.70

ollar

0.70

Activations Density 0.041%

top rankings or lists

references to ordinal rankings or lists

No Comments

No Known Activations