INDEX

Explanations

mentions or instances of the word "can" along with a following action or state

oai_token-act-pair · gpt-3.5-turbo

phrases expressing capability or potential

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

 rejection

-0.67

 Hunters

-0.64

bats

-0.61

 guarding

-0.60

 Rising

-0.59

 revision

-0.58

 Colony

-0.58

 Critics

-0.57

 danger

-0.57

mol

-0.56

POSITIVE LOGITS

't

1.53

berra

1.08

NOT

1.07

adian

1.05

¶ħ

0.93

isters

0.91

 afford

0.91

atell

0.89

 easily

0.89

 tell

0.88

Activations Density 0.117%

mentions or instances of the word "can" along with a following action or state

phrases expressing capability or potential

No Comments

No Known Activations