INDEX

Explanations

headline markers or delimiters like colons and brackets

oai_token-act-pair · gpt-3.5-turbo

colons and related punctuation indicating connections or references in text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

azo

-0.76

urat

-0.74

eatures

-0.74

ulz

-0.73

halla

-0.72

viation

-0.70

erver

-0.70

 grate

-0.68

 arrang

-0.67

ibble

-0.67

POSITIVE LOGITS

 Could

0.90

Why

0.88

 Woman

0.86

 Inside

0.85

 Seeking

0.85

 Beware

0.84

How

0.83

 Videos

0.83

 VIDEOS

0.82

 Picks

0.82

Activations Density 0.051%

headline markers or delimiters like colons and brackets

colons and related punctuation indicating connections or references in text

No Comments

No Known Activations