INDEX

Explanations

words related to broad or wide-ranging concepts or initiatives

oai_token-act-pair · gpt-3.5-turbo

the use of the term "wide" in various contexts referring to broad scope or extent

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

ilater

-0.79

anamo

-0.76

ulia

-0.72

Fight

-0.69

isons

-0.69

ngth

-0.68

ourke

-0.68

LV

-0.68

emis

-0.68

 Templ

-0.67

POSITIVE LOGITS

wide

1.12

sheets

0.81

 rollout

0.76

lining

0.74

sheet

0.73

 worldwide

0.73

 spectrum

0.72

 referen

0.72

 dissemination

0.71

 scope

0.70

Activations Density 0.008%

words related to broad or wide-ranging concepts or initiatives

the use of the term "wide" in various contexts referring to broad scope or extent

No Comments

No Known Activations