INDEX

Explanations

phrases related to decisions or intentions expressed in the future

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "would."

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

cius

-0.78

DAQ

-0.72

 sqor

-0.63

efeated

-0.61

RAM

-0.60

ception

-0.59

ancer

-0.58

chemy

-0.58

Lear

-0.57

 Parables

-0.57

POSITIVE LOGITS

ĸļ

0.90

 never

0.89

 gladly

0.84

be

0.83

 eventually

0.79

 soon

0.77

 reopen

0.74

 donate

0.74

 leave

0.74

aido

0.73

Activations Density 0.164%

phrases related to decisions or intentions expressed in the future

occurrences of the word "would."

No Comments

No Known Activations