INDEX

Explanations

words related to rodents, particularly rats

oai_token-act-pair · gpt-3.5-turbo

occurrences and variations of the word "rat."

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

 Mandela

-0.70

conservancy

-0.69

 Marketplace

-0.66

alach

-0.66

«ĺ

-0.65

OUS

-0.65

 Rica

-0.64

ELD

-0.61

 Hasan

-0.61

 Papua

-0.60

POSITIVE LOGITS

chet

1.69

che

1.20

cliffe

1.10

ting

1.08

atos

1.02

ifiers

1.01

ty

0.97

chery

0.96

nik

0.94

ches

0.93

Activations Density 0.018%

words related to rodents, particularly rats

occurrences and variations of the word "rat."

No Comments

No Known Activations