INDEX

Explanations

instances of threatening actions or statements

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "to" followed by various verbs indicating threats or promises

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

quickShipAvailable

-0.66

IED

-0.63

 Prediction

-0.63

puted

-0.61

pointer

-0.60

ortunate

-0.60

FactoryReloaded

-0.60

 portals

-0.60

points

-0.59

pointers

-0.58

POSITIVE LOGITS

 revive

1.02

 dismantle

1.01

 donate

0.99

 destro

0.99

 give

0.96

 leave

0.96

 settle

0.96

 injure

0.95

 endorse

0.94

 abandon

0.94

Activations Density 0.062%

instances of threatening actions or statements

occurrences of the word "to" followed by various verbs indicating threats or promises

No Comments

No Known Activations