INDEX

Explanations

mentions of the word "Snow" with varying activations

oai_token-act-pair · gpt-3.5-turbo

references to "Snow" in various contexts

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

igious

-0.83

ernandez

-0.73

ributes

-0.72

ulhu

-0.72

opathy

-0.70

ented

-0.69

ect

-0.68

arians

-0.67

izabeth

-0.67

icals

-0.66

POSITIVE LOGITS

flake

1.42

 Leopard

0.94

 Snow

0.93

don

0.87

dale

0.86

hawk

0.84

bottom

0.83

den

0.83

tro

0.82

wind

0.82

Activations Density 0.013%

mentions of the word "Snow" with varying activations

references to "Snow" in various contexts

No Comments

No Known Activations