INDEX

Explanations

mentions of physical holes

oai_token-act-pair · gpt-3.5-turbo

instances of the word "hole"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

cia

-0.70

Bal

-0.63

uscript

-0.63

 livest

-0.61

lux

-0.59

ority

-0.59

nesota

-0.58

 Balanced

-0.58

 normalized

-0.57

 Idle

-0.57

POSITIVE LOGITS

 hole

1.25

 Hole

1.20

 holes

1.16

hole

1.10

izons

0.98

pots

0.93

 drilled

0.90

holes

0.90

rition

0.86

dug

0.82

Activations Density 0.021%

mentions of physical holes

instances of the word "hole"

No Comments

No Known Activations