INDEX

Explanations

phrases containing strong negative or critical language

oai_token-act-pair · gpt-3.5-turbo

phrases that emphasize the concept of a "total" or complete negative experience

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

maid

-0.84

hops

-0.76

yer

-0.73

 Carbuncle

-0.72

bees

-0.70

pell

-0.68

lyn

-0.67

 cider

-0.67

utics

-0.66

paces

-0.66

POSITIVE LOGITS

itarian

1.16

itar

0.89

 strangers

0.79

 annihilation

0.74

ization

0.69

 domination

0.69

iza

0.67

 coincidence

0.66

iosity

0.66

 effort

0.66

Activations Density 0.020%

phrases containing strong negative or critical language

phrases that emphasize the concept of a "total" or complete negative experience

No Comments

No Known Activations