INDEX

Explanations

profane and offensive language

oai_token-act-pair · gpt-3.5-turbo

words related to bodily actions and expressions, often with a humorous or vulgar tone

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

eting

-0.69

eters

-0.64

 subcontract

-0.63

ipers

-0.62

eder

-0.59

 Guards

-0.59

ļéĨĴ

-0.58

rence

-0.57

atu

-0.57

 ration

-0.56

POSITIVE LOGITS

vana

0.82

 lucky

0.71

bit

0.69

licks

0.68

 darn

0.67

THING

0.66

 messy

0.66

 unlucky

0.64

anked

0.64

 luck

0.63

Activations Density 0.316%

profane and offensive language

words related to bodily actions and expressions, often with a humorous or vulgar tone

No Comments

No Known Activations