INDEX

Explanations

instances of the word "can" followed by a verb in past tense

oai_token-act-pair · gpt-3.5-turbo

instances of the word "can," indicating potential or capability

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

edient

-0.68

 Mant

-0.63

 revision

-0.62

furt

-0.62

 Fighter

-0.62

 Strikes

-0.60

 guarding

-0.60

 rejection

-0.58

 mindset

-0.57

 Gadget

-0.57

POSITIVE LOGITS

't

1.60

berra

1.29

adian

1.24

NOT

1.03

tera

0.94

 afford

0.90

 attest

0.88

isters

0.87

vas

0.87

aden

0.87

Activations Density 0.167%

instances of the word "can" followed by a verb in past tense

instances of the word "can," indicating potential or capability

No Comments

No Known Activations