INDEX

Explanations

phrases related to actions or statuses of people

oai_token-act-pair · gpt-3.5-turbo

emotions and states of conflict in narratives

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

 respectively

-0.73

?".

-0.72

 selves

-0.70

 thereof

-0.69

selves

-0.66

$.

-0.65

 collectively

-0.64

VERTISEMENT

-0.64

ĨĴ

-0.64

}.

-0.63

POSITIVE LOGITS

 himself

0.81

itone

0.77

 herself

0.69

veland

0.67

his

0.64

ansky

0.64

hirt

0.63

isconsin

0.56

agle

0.56

essen

0.56

Activations Density 1.283%

phrases related to actions or statuses of people

emotions and states of conflict in narratives

No Comments

No Known Activations