INDEX

Explanations

the word "te" followed by a number that is likely referring to different levels or values of something

oai_token-act-pair · gpt-3.5-turbo

occurrences of the syllable 'te'

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 demographics

-0.64

 Ridley

-0.63

 Timberwolves

-0.62

INGTON

-0.60

owsky

-0.60

 Icelandic

-0.60

OV

-0.59

 rounding

-0.57

 minded

-0.56

eous

-0.56

POSITIVE LOGITS

pee

1.32

legraph

1.31

legram

1.28

pees

1.11

eming

1.06

achable

1.04

cn

1.03

quila

1.02

asers

0.99

ak

0.87

Activations Density 0.019%

the word "te" followed by a number that is likely referring to different levels or values of something

occurrences of the syllable 'te'

No Comments

No Known Activations