INDEX

Explanations

terms indicating a lack of hesitation or reluctance

oai_token-act-pair · gpt-3.5-turbo

terms related to hesitation

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

arov

-0.87

ham

-0.82

toc

-0.78

elf

-0.78

rollers

-0.77

monds

-0.77

quality

-0.77

model

-0.75

rix

-0.72

andals

-0.69

POSITIVE LOGITS

 hesitation

1.06

 hesitate

0.99

 hesitated

0.89

hes

0.83

 hesitant

0.76

 reluctance

0.73

theless

0.71

 apprehension

0.70

UGE

0.70

 reluct

0.68

Activations Density 0.020%

terms indicating a lack of hesitation or reluctance

terms related to hesitation

No Comments

No Known Activations