INDEX

Explanations

comparative adjectives with negative connotations

oai_token-act-pair · gpt-3.5-turbo

phrases that express escalating negative emotions or concerns

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

#$

-0.85

bern

-0.71

uum

-0.70

 decaying

-0.61

icide

-0.61

cking

-0.58

uel

-0.58

ioxide

-0.56

olon

-0.56

 Fires

-0.55

POSITIVE LOGITS

 importantly

0.80

orthy

0.76

Ø©

0.71

 though

0.68

ici

0.66

 however

0.66

nown

0.65

ANG

0.63

 note

0.63

quickShipAvailable

0.62

Activations Density 0.157%

comparative adjectives with negative connotations

phrases that express escalating negative emotions or concerns

No Comments

No Known Activations