INDEX

Explanations

adjectives describing things that are distinguishably dissimilar or atypical

oai_token-act-pair · gpt-3.5-turbo

instances of the word "different" to highlight varying situations or contexts

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

RF

-0.64

APH

-0.63

hers

-0.61

âĢł

-0.60

 erection

-0.59

FL

-0.58

rollers

-0.56

WI

-0.56

liner

-0.56

POSITIVE LOGITS

iating

1.61

iates

1.44

iated

1.19

iator

1.16

iate

1.06

iations

1.02

ials

1.01

ially

1.00

iable

0.99

iation

0.93

Activations Density 0.036%

adjectives describing things that are distinguishably dissimilar or atypical

instances of the word "different" to highlight varying situations or contexts

No Comments

No Known Activations