INDEX

Explanations

instances of the word "there" with high activation values

oai_token-act-pair · gpt-3.5-turbo

the word "there" and its variations 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 @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

rouse

-0.66

ize

-0.64

resses

-0.60

 Fortune

-0.55

IZE

-0.55

ave

-0.54

iling

-0.54

 Travels

-0.54

 PLUS

-0.53

pur

-0.53

POSITIVE LOGITS

 there

2.82

there

2.26

 THERE

2.25

 There

1.53

There

1.40

 therein

1.20

 here

1.13

 thence

1.01

 somewhere

0.99

 they

0.91

Activations Density 0.124%

instances of the word "there" with high activation values

the word "there" and its variations in various contexts

No Comments

No Known Activations