INDEX

Explanations

phrases related to actions carried out by individuals or groups of people

oai_token-act-pair · gpt-3.5-turbo

past tense verbs indicating significant actions or events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

():

-0.70

veland

-0.63

mania

-0.61

iac

-0.61

yond

-0.59

angs

-0.58

omorph

-0.58

KY

-0.58

avin

-0.57

xual

-0.57

POSITIVE LOGITS

 themselves

1.14

 respectively

1.02

 unanimously

0.85

 collectively

0.83

 their

0.79

 respective

0.74

 alike

0.74

 unison

0.71

 jointly

0.71

 together

0.71

Activations Density 0.617%

phrases related to actions carried out by individuals or groups of people

past tense verbs indicating significant actions or events

No Comments

No Known Activations