INDEX

Explanations

sets of related words, particularly noun phrases with multiple components

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate quantities or collections of items

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

hyde

-0.82

uca

-0.76

cture

-0.75

matter

-0.74

ivo

-0.74

ment

-0.74

eri

-0.72

ablishment

-0.72

terness

-0.72

ctuary

-0.71

POSITIVE LOGITS

 stairs

1.03

 interconnected

1.02

 essays

0.93

 events

0.91

 posts

0.88

 twins

0.88

 tweets

0.87

 handcuffs

0.87

 equations

0.86

 letters

0.84

Activations Density 0.081%

sets of related words, particularly noun phrases with multiple components

phrases that indicate quantities or collections of items

No Comments

No Known Activations