INDEX

Explanations

instances of the word "be" with a high importance placed on the number 9

oai_token-act-pair · gpt-3.5-turbo

phrases indicating negation or the inability to perform an action

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

 brace

-0.69

 traverse

-0.69

 Might

-0.66

 srfAttach

-0.66

 compose

-0.65

 showc

-0.63

nod

-0.62

strous

-0.62

iop

-0.62

 defy

-0.62

POSITIVE LOGITS

 anymore

1.14

 able

1.09

 bothered

1.02

 necessarily

0.96

 counted

0.93

 forgiven

0.93

 harmed

0.90

 anywhere

0.88

yet

0.85

 entirely

0.84

Activations Density 0.102%

instances of the word "be" with a high importance placed on the number 9

phrases indicating negation or the inability to perform an action

No Comments

No Known Activations