INDEX

Explanations

words related to amazement, astonishment, and admiration

oai_token-act-pair · gpt-3.5-turbo

expressions of admiration and wonder

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

sol

-0.67

å¥³

-0.66

Interstitial

-0.64

 Obesity

-0.63

 Thieves

-0.63

individual

-0.63

arians

-0.61

 Agents

-0.60

Solid

-0.59

Opt

-0.59

POSITIVE LOGITS

awe

1.01

urous

0.93

ingly

0.91

eful

0.88

htaking

0.88

efully

0.86

ruciating

0.79

fulness

0.77

 amaz

0.75

 aback

0.74

Activations Density 0.010%

words related to amazement, astonishment, and admiration

expressions of admiration and wonder

No Comments

No Known Activations