INDEX

Explanations

phrases related to mental states or perceptions about oneself or others

oai_token-act-pair · gpt-3.5-turbo

present tense and past tense forms of the verb "to be."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

avascript

-0.76

urry

-0.75

aughters

-0.70

aturdays

-0.70

inav

-0.68

 happens

-0.67

"],

-0.67

brates

-0.66

 Moines

-0.65

idth

-0.65

POSITIVE LOGITS

 capable

1.09

 destined

1.08

 willing

1.07

 going

1.02

 indeed

0.99

nt

0.99

 ready

0.99

 hiding

0.97

 able

0.97

 behaving

0.96

Activations Density 0.331%

phrases related to mental states or perceptions about oneself or others

present tense and past tense forms of the verb "to be."

No Comments

No Known Activations