INDEX

Explanations

phrases indicating evidence or examples supporting a stated proposition

oai_token-act-pair · gpt-3.5-turbo

instances of the word "Such" indicating a pattern or emphasis in the text

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

oil

-0.69

kick

-0.68

osph

-0.64

Ö¼

-0.64

 Polo

-0.63

creen

-0.63

olate

-0.62

 Loaded

-0.61

office

-0.61

iliate

-0.61

POSITIVE LOGITS

ities

0.81

 matters

0.79

ties

0.79

 embodiments

0.67

 discrepancies

0.67

cond

0.64

 cancell

0.63

 phenomena

0.63

 Takeru

0.63

fter

0.63

Activations Density 0.035%

phrases indicating evidence or examples supporting a stated proposition

instances of the word "Such" indicating a pattern or emphasis in the text

No Comments

No Known Activations