INDEX

Explanations

profane or strong language

oai_token-act-pair · gpt-3.5-turbo

references to curses or swearing

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

å§«

-1.00

nington

-0.81

atican

-0.80

parency

-0.78

issance

-0.77

arnaev

-0.76

olitan

-0.76

oulos

-0.76

anooga

-0.75

itutional

-0.74

POSITIVE LOGITS

 curse

0.91

words

0.84

 curses

0.84

 words

0.78

hammer

0.78

 cursing

0.75

 cursed

0.73

bones

0.70

word

0.67

vine

0.67

Activations Density 0.036%

profane or strong language

references to curses or swearing

No Comments

No Known Activations